High % shots - come on now - we all know where those come from:
The problem with defining high-percentage shots solely in terms of location (the home-plate area) is that location isn't anywhere near as important as movement and deception in determining whether a shot goes in. Even within the home-plate area, goalies save an incredibly high percentage of "clean shots" - unscreened shots where the goalie can square to the puck and set depth and angle with minimal lateral movement. Even shots from the hashmarks are saved 85-90% of the time, while shots from above the circles are saved 93-95% of the time:
If you want to beat NHL goalies, you have to screen them, deflect the shot, or, above all, get them moving side-to-side. Once you get a goalie moving side to side, the save percentage plummets - to less than 70%. (And that's on all transition shots - the save percentage on one-timers is even lower.) So the key factor is not where the shot is taken - it's what happens before the shot.
Realistic goalie tuning, then, would have one-timers being roughly six times as effective as wristers/slapshots from above the circles. If you play this game, you know that the tuning in this game is nothing like that, and hasn't been since at least NHL 12.
I'll just say the same thing I've been saying for years: goalies in this game should be worse at making saves when they have to move side-to-side, and better at saving unscreened shots from distance. That's both more true-to-life, and it would also encourage better hockey, since it would incentivize people to pass the puck rather than just circle around trying to get off a high-slot wrister.
In a perfect world, we could minimize a few of the weaker goals but when you are dealing with single frame differences, it becomes a very fine line. In a couple of the videos shown, the goalie was late by a frame or two on the screen and he was also late by a frame on the deflection reaction but just got a piece of it with his glove. This comes down to the speed of the shot on that one screen and the speed taken off during the deflection.
I don't get this. Yes, the deflection took a little speed off the puck, but it also completely altered its direction. Effectively, the goalie had to make a new save on a shot that was taken from 15 feet away, but that was traveling only slightly slower than the original slapshot. There's no way for him to make that save deliberately. Human reaction times just aren't fast enough.
You seem to be saying the code that determines the goalie's reaction to a deflection (in terms of how fast he reacts) is the same as the code that determines the goalie's reaction to a shot from distance. But there's no way they can be the same. If a goalie can react fast enough to deliberately save a deflected shot from 10-15 feet away, they would save every shot from 55 feet away with ease. Shots from distance would look like the bullets in The Matrix to them.
It's so painfully obvious short side goals are broken.
Almost half of the last 30 games I played was me saying to myself, here we go again, all I gotta do is just stand in front of this guy every time he enters the zone because the only thing he is attempting to do is go short side.
That is just boring hockey and ruins all the fun. You don't need to worry much anymore about one timers, rebounds or point shots for the tip. Almost every game is the same story, no team play, just enter zone try to cut in and go short side because it's more effective than cycling and making plays.
If that is all players are doing now that proves beyond all doubt that short side is broken and EA needs to fix it.
Should EA then program the game so that shots from a certain area are no longer allowed to cross the goal line, no matter the circumstances? Then you begin to chip away at the organic nature in which goals (for the most part) are generated.
We could also handle this organically under the same methodology that we currently tune the game but there are knock ons to that as well -- for example, people feeling they aren't rewarded enough for good chances or that the game feels more random. We already hear people say these things under the current tuning (in regards to missing the net or goalies robbing them) so there is already a skill gap in the execution and getting time and space to settle, etc.
If a goalies worst possible delay is decreased for example, you will see less goals from distance but since we are talking the difference of a frame in most cases, it probably means lowering the rate of all times the goalie is late by a frame so a lot of great chances in the slot are diminished as well.
So then when you get to a competitive game where a player doesn't give up any shots in close against an opponent that gives them two clean shots from the slot, stuck in a 0-0 game, where right now, you would hope at the worst it would be 2-1 or in most cases 2-0
If it were easy to tune goalies, we would have been done long ago but it is something we never stop looking at. Right now, we are looking at short side more than we are looking at the shots from the top of the circles or screen effects as for the most part those are defendable and encouraging manual defense to get a stick or body in to disrupt a shot is a good thing overall.
In a perfect world, we could minimize a few of the weaker goals but when you are dealing with single frame differences, it becomes a very fine line. In a couple of the videos shown, the goalie was late by a frame or two on the screen and he was also late by a frame on the deflection reaction but just got a piece of it with his glove. This comes down to the speed of the shot on that one screen and the speed taken off during the deflection. A few mph difference, or a little less screen and both those chances go the other way. As always, not making any excuses for it, just trying to shed some light into what is going on there and how fine of a line it can be. You move that fine line and a lot of things that are working for the game's balance then shift as well.
Appreciate seeing the examples and comments from everyone though as it allows us to take more perspectives than our own into consideration as we continue to look at things.
Have you tried lowering shot power and goalie reaction time to compensate?
The lowering of shot power in itself grants a few frames and could possibly give wiggle room on fine tuning reaction times for the goalie.
Have you tried lowering shot power and goalie reaction time to compensate?
The lowering of shot power in itself grants a few frames and could possibly give wiggle room on fine tuning reaction times for the goalie.
Although it is true that slower shots give more frames to play with, it still yields the same results overall as you are only talking about the difference in 4-5 hundredths of a second when talking about a shot from the point that is 10mph less for example. So you are still dealing with single frames at some stage.
Would also prefer not to compromise on the authenticity of real world speeds for shots either.
We could get something even more realistic with shot accuracy, goalie reaction times, etc., we just know that it starts to feel more random to people at times and that we aren't rewarding good opportunities enough when it comes to competitive game balance.
It is just a fine line but I like the way you are thinking.
Have you tried lowering shot power and goalie reaction time to compensate?
The lowering of shot power in itself grants a few frames and could possibly give wiggle room on fine tuning reaction times for the goalie.
Although it is true that slower shots give more frames to play with, it still yields the same results overall as you are only talking about the difference in 4-5 hundredths of a second when talking about a shot from the point that is 10mph less for example. So you are still dealing with single frames at some stage.
Would also prefer not to compromise on the authenticity of real world speeds for shots either.
We could get something even more realistic with shot accuracy, goalie reaction times, etc., we just know that it starts to feel more random to people at times and that we aren't rewarding good opportunities enough when it comes to competitive game balance.
It is just a fine line but I like the way you are thinking.
It's been a while since I've played offline but I can try tinkering with some stuff to see and I'll post a game or 2. Hopefully sometime soon and then you can have info on that.
I don't get this. Yes, the deflection took a little speed off the puck, but it also completely altered its direction. Effectively, the goalie had to make a new save on a shot that was taken from 15 feet away, but that was traveling only slightly slower than the original slapshot. There's no way for him to make that save deliberately. Human reaction times just aren't fast enough.
You seem to be saying the code that determines the goalie's reaction to a deflection (in terms of how fast he reacts) is the same as the code that determines the goalie's reaction to a shot from distance. But there's no way they can be the same. If a goalie can react fast enough to deliberately save a deflected shot from 10-15 feet away, they would save every shot from 55 feet away with ease. Shots from distance would look like the bullets in The Matrix to them.
There are a few things I can dive into but your thoughts also assume that they have the same read every time. A goalie doesn't react to a shot 15 feet away with the same reaction time every time as other factors weigh into it as well. They have error due to their ratings, it depends if they were settled first, how quickly they first saw where the shot was shot from, deflected from, etc., if they were thinking the player would shoot, pass or carry/deke the puck on regular shot attempts, etc.
And even after that, we are still talking the difference of a couple frames. A 45mph shot form 18 feet out to a 70 mph shot from the hashmarks to a slapshot from the point are only a difference of 1.4 tenths of a second and I would argue that our goalie would in most cases stop the open read at a slapshot from the point more than they would actively save a deflection from in close. However, in most cases they just get big and the deflection just hits them which is part of what a real world goalie does as well to increase their odds to not rely solely on reactionary saves when it will be tough on them.
If you got into practice mode, even though they may get hit incidentally with the puck, you will see they are actually late reacting to deflections far more than they are late to get a blocker or glove on an open slapshot from the point.
The problem with defining high-percentage shots solely in terms of location (the home-plate area) is that location isn't anywhere near as important as movement and deception in determining whether a shot goes in. Even within the home-plate area, goalies save an incredibly high percentage of "clean shots" - unscreened shots where the goalie can square to the puck and set depth and angle with minimal lateral movement. Even shots from the hashmarks are saved 85-90% of the time, while shots from above the circles are saved 93-95% of the time:
If you want to beat NHL goalies, you have to screen them, deflect the shot, or, above all, get them moving side-to-side. Once you get a goalie moving side to side, the save percentage plummets - to less than 70%. (And that's on all transition shots - the save percentage on one-timers is even lower.) So the key factor is not where the shot is taken - it's what happens before the shot.
Realistic goalie tuning, then, would have one-timers being roughly six times as effective as wristers/slapshots from above the circles. If you play this game, you know that the tuning in this game is nothing like that, and hasn't been since at least NHL 12.
I'll just say the same thing I've been saying for years: goalies in this game should be worse at making saves when they have to move side-to-side, and better at saving unscreened shots from distance. That's both more true-to-life, and it would also encourage better hockey, since it would incentivize people to pass the puck rather than just circle around trying to get off a high-slot wrister.
These are all great stats and a really good way to look at it. That said, shot location does matter as it determines the time a goalie has to react and the angle changes how much of the net they can block just by being big before they need to react to make the save.
On top of that, the quality of the shot (power and accuracy) cross over with that as well. If and NHL player puts it into the goalies chest, he will make the initial save hopefully as easily regardless of the position on the ice. If our shot accuracy is elevated in the best case scenarios it moves those percentages as well and we know in our best case scenario, shot accuracy probably does have less spray more often compared to the real world (even though our error at it's worst is fairly accurate) but that is to remove that feeling of randomness and works as long as the defensive tools make it hard for a player to get settled and shoot uncontested -- which is obviously another subjective piece but both teams are playing under the same mechanics.
As with anything it is all a balance and I think you are looking at the right things but everything is a bit heightened and more forgiving on the Competitive Preset that online uses relative to what we would tune if we were playing 20 minute periods and full realism.
I actually think our one timers do score at a way higher rate than shots from those positions on the ice when just taking general samples and I do think that if you just took the sample size of NHL players that had time to settle and get a perfect shot from those areas of the ice, you would see those numbers increase by a large amount as well.
The problem with defining high-percentage shots solely in terms of location (the home-plate area) is that location isn't anywhere near as important as movement and deception in determining whether a shot goes in. Even within the home-plate area, goalies save an incredibly high percentage of "clean shots" - unscreened shots where the goalie can square to the puck and set depth and angle with minimal lateral movement. Even shots from the hashmarks are saved 85-90% of the time, while shots from above the circles are saved 93-95% of the time:
If you want to beat NHL goalies, you have to screen them, deflect the shot, or, above all, get them moving side-to-side. Once you get a goalie moving side to side, the save percentage plummets - to less than 70%. (And that's on all transition shots - the save percentage on one-timers is even lower.) So the key factor is not where the shot is taken - it's what happens before the shot.
Realistic goalie tuning, then, would have one-timers being roughly six times as effective as wristers/slapshots from above the circles. If you play this game, you know that the tuning in this game is nothing like that, and hasn't been since at least NHL 12.
I'll just say the same thing I've been saying for years: goalies in this game should be worse at making saves when they have to move side-to-side, and better at saving unscreened shots from distance. That's both more true-to-life, and it would also encourage better hockey, since it would incentivize people to pass the puck rather than just circle around trying to get off a high-slot wrister.
These are all great stats and a really good way to look at it. That said, shot location does matter as it determines the time a goalie has to react and the angle changes how much of the net they can block just by being big before they need to react to make the save.
On top of that, the quality of the shot (power and accuracy) cross over with that as well. If and NHL player puts it into the goalies chest, he will make the initial save hopefully as easily regardless of the position on the ice. If our shot accuracy is elevated in the best case scenarios it moves those percentages as well and we know in our best case scenario, shot accuracy probably does have less spray more often compared to the real world (even though our error at it's worst is fairly accurate) but that is to remove that feeling of randomness and works as long as the defensive tools make it hard for a player to get settled and shoot uncontested -- which is obviously another subjective piece but both teams are playing under the same mechanics.
As with anything it is all a balance and I think you are looking at the right things but everything is a bit heightened and more forgiving on the Competitive Preset that online uses relative to what we would tune if we were playing 20 minute periods and full realism.
I actually think our one timers do score at a way higher rate than shots from those positions on the ice when just taking general samples and I do think that if you just took the sample size of NHL players that had time to settle and get a perfect shot from those areas of the ice, you would see those numbers increase by a large amount as well.
I agree with most of what you say except for the one timers part. I feel that in the middle of the slot, the goalies take away so much net by the time the puck gets to that area that it's tough to score. You almost have to be across the crease in order to score this year on one timers. If you put it out in the middle slot from the side it almost seems like the goalies get there before the puck and then take away a vast majority of the net already and when you think you can go far side the body is already there.
But if you were to shoot from that area with possession it seems there's more net to shoot at.
I agree with most of what you say except for the one timers part. I feel that in the middle of the slot, the goalies take away so much net by the time the puck gets to that area that it's tough to score. You almost have to be across the crease in order to score this year on one timers. If you put it out in the middle slot from the side it almost seems like the goalies get there before the puck and then take away a vast majority of the net already and when you think you can go far side the body is already there.
But if you were to shoot from that area with possession it seems there's more net to shoot at.
It is more of a shot accuracy thing in those cases. The goalie is less precise in his saves and his timing is slower to react when he isn't settled to see the release of the shot. So on one timers, he often does slide over more blindly to take away as much net as possible and then reacts after that often with less success than if he was settled to see the shot release the whole time.
When you are in the middle of the slot to shoot, you are going to be more accurate but the goalie is also settled. However, the reaction time is the goalies biggest enemy in this case. The harder part though on the offensive side is the shooter being open in that area long enough to settle to get that accuracy. Puck movement for a one timer, as long as the lane was open doesn't give the defense as much time to disrupt that shot.
In most cases with one timers, getting the puck into the top half of the net where the goalie isn't will yield a goal but you are right, the less the goalie has to move over, the higher his chance is to cover more net as he doesn't need to stretch out as much like he would trying to take away the lower half of the far side.
How much the goalie commits to the original puck carrier also weighs as a big factor in the goalies ability to read and react quickly to the pass or not. If that puck carrier is skating in with his puck to his backhand curling away from the net as a non threat to score, the goalie is more likely to be anticipating pass and will get over much quicker than if the player is threat to shoot and you get the goalie to commit with some momentum before making the pass against it.
From my experience the reason 1 timers work less often in the slot than wrist shots has nothing to do with accuracy. It's about the animations. If you receive a pass and try to one time it "far side" there is a built in reaction animation that just makes that anticipates and makes that save. If you just skate in and take a wrister, and you know what you are doing, you are essentially exploiting the mechanics that tells the goalie to go down in butterfly even though he shouldn't in that situation. When he does this the corners and over the shoulders open wide. It's all about learning the goalie animations and knowing what they will do in specific scenarios.
Best way to score, besides just walking in and exploiting the butterfly goalie animations is to be wide for sharp angle shots. There's sweet spots. Again it's about learning the animations and exploiting them. Thats why 3v3 is simply about ragging the puck to draw penalties so you can get penalty shots and do bh/fh/bh all day for 100% goal scoring.
Just an aside: count me in when it comes to extending 6v6 drop in’s by 1-3 min a period(or more!). So often it feels as if the chemistry is just starting to gel late in the 2nd or early third and then it’s over..
All Comments pertain to 6v6 drop in unless otherwise stated..
I agree with most of what you say except for the one timers part. I feel that in the middle of the slot, the goalies take away so much net by the time the puck gets to that area that it's tough to score. You almost have to be across the crease in order to score this year on one timers. If you put it out in the middle slot from the side it almost seems like the goalies get there before the puck and then take away a vast majority of the net already and when you think you can go far side the body is already there.
But if you were to shoot from that area with possession it seems there's more net to shoot at.
It is more of a shot accuracy thing in those cases. The goalie is less precise in his saves and his timing is slower to react when he isn't settled to see the release of the shot. So on one timers, he often does slide over more blindly to take away as much net as possible and then reacts after that often with less success than if he was settled to see the shot release the whole time.
When you are in the middle of the slot to shoot, you are going to be more accurate but the goalie is also settled. However, the reaction time is the goalies biggest enemy in this case. The harder part though on the offensive side is the shooter being open in that area long enough to settle to get that accuracy. Puck movement for a one timer, as long as the lane was open doesn't give the defense as much time to disrupt that shot.
In most cases with one timers, getting the puck into the top half of the net where the goalie isn't will yield a goal but you are right, the less the goalie has to move over, the higher his chance is to cover more net as he doesn't need to stretch out as much like he would trying to take away the lower half of the far side.
How much the goalie commits to the original puck carrier also weighs as a big factor in the goalies ability to read and react quickly to the pass or not. If that puck carrier is skating in with his puck to his backhand curling away from the net as a non threat to score, the goalie is more likely to be anticipating pass and will get over much quicker than if the player is threat to shoot and you get the goalie to commit with some momentum before making the pass against it.
I'm gonna look at more video to investigate some stuff. Because I feel that their reaction is maybe too high from one timers? I'm not sure if that's something that can be modified with your system or not but I'm going to look at and see if the shots from the middle slot are being gloved/blockered as opposed to bodied.
I'm gonna look at more video to investigate some stuff. Because I feel that their reaction is maybe too high from one timers? I'm not sure if that's something that can be modified with your system or not but I'm going to look at and see if the shots from the middle slot are being gloved/blockered as opposed to bodied.
They will sometimes be saved with reactionary saves with their limbs as well. I am just saying that overall, their reaction time and precision will be slower the more they have to move and the less settled they are. It isn't black and white that you will never see it. Goalies make great glove saves on one timers in the real world too but those are more incredible saves than glove saves on longer weaker shots they are able to see easily.
From my experience the reason 1 timers work less often in the slot than wrist shots has nothing to do with accuracy. It's about the animations. If you receive a pass and try to one time it "far side" there is a built in reaction animation that just makes that anticipates and makes that save. If you just skate in and take a wrister, and you know what you are doing, you are essentially exploiting the mechanics that tells the goalie to go down in butterfly even though he shouldn't in that situation. When he does this the corners and over the shoulders open wide. It's all about learning the goalie animations and knowing what they will do in specific scenarios.
Best way to score, besides just walking in and exploiting the butterfly goalie animations is to be wide for sharp angle shots. There's sweet spots. Again it's about learning the animations and exploiting them. Thats why 3v3 is simply about ragging the puck to draw penalties so you can get penalty shots and do bh/fh/bh all day for 100% goal scoring.
Goalies often drop on shots and then have to react to shots that are away from their body and they are more vulnerable the more central the shooter is and the more net there is to cover. It forces a good shot to beat them and then their reaction time gives a chance to stop a better shot after that.
On one timers, goalies will blindly/pre-emptively slide over and take away as much net as they can after they read the pass. On a wrist shot, that is essentially getting big in butterfly and the better the angle of the shot, the more net you have to shoot at. A goalie tries to get their positioning right to take away as much net as possible. On one timers, the side of the net they need to get to is more obvious so they can commit with their leg and arm to that side but that is why they also get beat against the grain or easily up and over their leg/arm into a bigger area of the net than is usually there with a good angle cut off on a regular wrist shot -- but in either case, a great shot can beat the goalie if they can't react in time which is why accuracy is a big piece to the puzzle as well.
I'm gonna look at more video to investigate some stuff. Because I feel that their reaction is maybe too high from one timers? I'm not sure if that's something that can be modified with your system or not but I'm going to look at and see if the shots from the middle slot are being gloved/blockered as opposed to bodied.
They will sometimes be saved with reactionary saves with their limbs as well. I am just saying that overall, their reaction time and precision will be slower the more they have to move and the less settled they are. It isn't black and white that you will never see it. Goalies make great glove saves on one timers in the real world too but those are more incredible saves than glove saves on longer weaker shots they are able to see easily.
I understand what you're saying, but being that close to the net, there shouldn't be enough time to react unless they're just getting a piece of it really. Most goalies throw themselves into position on one timers and hope the puck hits them, react secondary. I'm just saying often the goalies are reacting too quickly for the one timers, but not reacting enough sometimes for shots at distances.
I believe you saw that video with the tip in front where the goalie who has no chance somehow reacts with the glove. Can't remember which thread it's in. You see those, and it almost feels as if the game is pre programmed to make saves. It's likely not that way, but if you saw those saves against you and your goalie is allowing clappers far side from 5 feet off the top of the circle you could use your imagination and almost feel as if the game is fixed.
There are a few things I can dive into but your thoughts also assume that they have the same read every time. A goalie doesn't react to a shot 15 feet away with the same reaction time every time as other factors weigh into it as well.
I would argue that our goalie would in most cases stop the open read at a slapshot from the point more than they would actively save a deflection from in close.
Right, but my point is that the two things aren't even in the same ballpark. Even if it's a 95-mph shot, if it's an unscreened shot taken from 55 feet away, that gives the goalie .40 seconds to react and make the save. That's more than enough time for an NHL goalie to make a save, especially given that he has a clear view of the shooter and so can read the puck right off the stick.
With a deflection off a slapshot, there is essentially no time to make a reaction save. If the guy (as in the clip below) is 16-18 feet away (let alone closer), even with the deflection slowing the puck the goalie's going to have .20 seconds or less to react. That's close to the minimum possible reaction time for a human (even a well-trained human) to even start to move. And the goalie can't anticipate, because by definition a deflection can go anywhere. So if goalies' reactions in this game are good enough to make a deliberate save like this, where the goalie tracks the deflection and reaches up to knock the puck away (rather than just making himself big and getting lucky):
they should basically never be beaten from distance on unscreened shots, unless you've coded the game so goalies are in effect allowed to cheat on deflections (by giving them a heads-up about where the puck is going).
This is basically the same thing HoodHoppers just said in his post: goalies' reaction times on deflections and close-in one-timers seems totally different (as in, much faster) than their reaction times to above-the-circle shots.
Right, but my point is that the two things aren't even in the same ballpark. Even if it's a 95-mph shot, if it's an unscreened shot taken from 55 feet away, that gives the goalie .40 seconds to react and make the save. That's more than enough time for an NHL goalie to make a save, especially given that he has a clear view of the shooter and so can read the puck right off the stick.
With a deflection off a slapshot, there is essentially no time to make a reaction save. If the guy (as in the clip below) is 16-18 feet away (let alone closer), even with the deflection slowing the puck the goalie's going to have .20 seconds or less to react. That's close to the minimum possible reaction time for a human (even a well-trained human) to even start to move. And the goalie can't anticipate, because by definition a deflection can go anywhere. So if goalies' reactions in this game are good enough to make a deliberate save like this, where the goalie tracks the deflection and reaches up to knock the puck away (rather than just making himself big and getting lucky):
they should basically never be beaten from distance on unscreened shots, unless you've coded the game so goalies are in effect allowed to cheat on deflections (by giving them a heads-up about where the puck is going).
This is basically the same thing HoodHoppers just said in his post: goalies' reaction times on deflections and close-in one-timers seems totally different (as in, much faster) than their reaction times to above-the-circle shots.
You actually responded to my post that explains why this is possible but I should have explained further. The goalies don't react the exact same on every single shot. There is human error built in which is what their attributes factor in as well as other factors in terms of what is happening in front of them. They are constantly analyzing if they think the player is going to shoot, pass or skate/deke with the puck, they are recognizing other threats around them and physical factors such as their movement/momentum and time to move their body/limb to the puck play into it as well.
Again, I am glad that you look into things that detailed. Before I responded to you the first time I had actually broken down similar scenarios to have some math done on it as well regarding a deflection coming in at 45mph from 18-22 feet out vs a wrist shot of 70ish mph from the top of the circles vs a slapshot of 90mph from 55 feet out similar to what you did and it really comes down to a difference between .3 and .4 seconds. I don't think you are compensating for enough speed being taken off when a puck is deflected by as much in that example but I don't think that is the main thing to worry about as we are talking relatively similar numbers and it is all happening very quick with little room for error on either side. I had the luxury to try a few deflections in our test bed where I can look at the speeds to see as well and feel 45mph is actually generous. (I couldn't remember if I posted anything about this or had just thought about it so apologies if it repeats anything)
The difference of 1/10 of a second more delay before he starts to try and make his save or not can allow the slapshot to score in case with the same save reaction that he responds to right away on a deflection with that speed/distance. That also assumes the puck location relative to the goalie is the same as well. A deflection in our game would have more accuracy to the perimeter of the net more often than a deflection would.
In most cases the goalie will have more delay and won't be reacting right on point so shots that get to him quicker have a higher chance to score in all cases than those that don't when all is the same but at his best, he can still react quickly with a good read and it just depends how long it takes to get his limb to the puck to get a piece of it. A bit more delay in the deflection case and it's in the net and you can take plenty of slapshots from the blueline and have the goalie save them all day. So overall, it aligns with what you are saying.
And you are right, for a perfect simulation, maybe our min reaction times are too low, maybe our max delay is too high, maybe our max shot accuracy is too high but we have always said that on competitive game style everything is a bit heightened for 4 minute periods and to feel balanced and not random. It still leans quite a bit to the sim side as far as our game styles are concerned though.
Replies
The problem with defining high-percentage shots solely in terms of location (the home-plate area) is that location isn't anywhere near as important as movement and deception in determining whether a shot goes in. Even within the home-plate area, goalies save an incredibly high percentage of "clean shots" - unscreened shots where the goalie can square to the puck and set depth and angle with minimal lateral movement. Even shots from the hashmarks are saved 85-90% of the time, while shots from above the circles are saved 93-95% of the time:
https://www.sportsnet.ca/hockey/introducing-the-shot-quality-project-part-ii/
If you want to beat NHL goalies, you have to screen them, deflect the shot, or, above all, get them moving side-to-side. Once you get a goalie moving side to side, the save percentage plummets - to less than 70%. (And that's on all transition shots - the save percentage on one-timers is even lower.) So the key factor is not where the shot is taken - it's what happens before the shot.
Realistic goalie tuning, then, would have one-timers being roughly six times as effective as wristers/slapshots from above the circles. If you play this game, you know that the tuning in this game is nothing like that, and hasn't been since at least NHL 12.
I'll just say the same thing I've been saying for years: goalies in this game should be worse at making saves when they have to move side-to-side, and better at saving unscreened shots from distance. That's both more true-to-life, and it would also encourage better hockey, since it would incentivize people to pass the puck rather than just circle around trying to get off a high-slot wrister.
I don't get this. Yes, the deflection took a little speed off the puck, but it also completely altered its direction. Effectively, the goalie had to make a new save on a shot that was taken from 15 feet away, but that was traveling only slightly slower than the original slapshot. There's no way for him to make that save deliberately. Human reaction times just aren't fast enough.
You seem to be saying the code that determines the goalie's reaction to a deflection (in terms of how fast he reacts) is the same as the code that determines the goalie's reaction to a shot from distance. But there's no way they can be the same. If a goalie can react fast enough to deliberately save a deflected shot from 10-15 feet away, they would save every shot from 55 feet away with ease. Shots from distance would look like the bullets in The Matrix to them.
Almost half of the last 30 games I played was me saying to myself, here we go again, all I gotta do is just stand in front of this guy every time he enters the zone because the only thing he is attempting to do is go short side.
That is just boring hockey and ruins all the fun. You don't need to worry much anymore about one timers, rebounds or point shots for the tip. Almost every game is the same story, no team play, just enter zone try to cut in and go short side because it's more effective than cycling and making plays.
If that is all players are doing now that proves beyond all doubt that short side is broken and EA needs to fix it.
Have you tried lowering shot power and goalie reaction time to compensate?
The lowering of shot power in itself grants a few frames and could possibly give wiggle room on fine tuning reaction times for the goalie.
Although it is true that slower shots give more frames to play with, it still yields the same results overall as you are only talking about the difference in 4-5 hundredths of a second when talking about a shot from the point that is 10mph less for example. So you are still dealing with single frames at some stage.
Would also prefer not to compromise on the authenticity of real world speeds for shots either.
We could get something even more realistic with shot accuracy, goalie reaction times, etc., we just know that it starts to feel more random to people at times and that we aren't rewarding good opportunities enough when it comes to competitive game balance.
It is just a fine line but I like the way you are thinking.
It's been a while since I've played offline but I can try tinkering with some stuff to see and I'll post a game or 2. Hopefully sometime soon and then you can have info on that.
There are a few things I can dive into but your thoughts also assume that they have the same read every time. A goalie doesn't react to a shot 15 feet away with the same reaction time every time as other factors weigh into it as well. They have error due to their ratings, it depends if they were settled first, how quickly they first saw where the shot was shot from, deflected from, etc., if they were thinking the player would shoot, pass or carry/deke the puck on regular shot attempts, etc.
And even after that, we are still talking the difference of a couple frames. A 45mph shot form 18 feet out to a 70 mph shot from the hashmarks to a slapshot from the point are only a difference of 1.4 tenths of a second and I would argue that our goalie would in most cases stop the open read at a slapshot from the point more than they would actively save a deflection from in close. However, in most cases they just get big and the deflection just hits them which is part of what a real world goalie does as well to increase their odds to not rely solely on reactionary saves when it will be tough on them.
If you got into practice mode, even though they may get hit incidentally with the puck, you will see they are actually late reacting to deflections far more than they are late to get a blocker or glove on an open slapshot from the point.
On top of that, the quality of the shot (power and accuracy) cross over with that as well. If and NHL player puts it into the goalies chest, he will make the initial save hopefully as easily regardless of the position on the ice. If our shot accuracy is elevated in the best case scenarios it moves those percentages as well and we know in our best case scenario, shot accuracy probably does have less spray more often compared to the real world (even though our error at it's worst is fairly accurate) but that is to remove that feeling of randomness and works as long as the defensive tools make it hard for a player to get settled and shoot uncontested -- which is obviously another subjective piece but both teams are playing under the same mechanics.
As with anything it is all a balance and I think you are looking at the right things but everything is a bit heightened and more forgiving on the Competitive Preset that online uses relative to what we would tune if we were playing 20 minute periods and full realism.
I actually think our one timers do score at a way higher rate than shots from those positions on the ice when just taking general samples and I do think that if you just took the sample size of NHL players that had time to settle and get a perfect shot from those areas of the ice, you would see those numbers increase by a large amount as well.
I agree with most of what you say except for the one timers part. I feel that in the middle of the slot, the goalies take away so much net by the time the puck gets to that area that it's tough to score. You almost have to be across the crease in order to score this year on one timers. If you put it out in the middle slot from the side it almost seems like the goalies get there before the puck and then take away a vast majority of the net already and when you think you can go far side the body is already there.
But if you were to shoot from that area with possession it seems there's more net to shoot at.
When you are in the middle of the slot to shoot, you are going to be more accurate but the goalie is also settled. However, the reaction time is the goalies biggest enemy in this case. The harder part though on the offensive side is the shooter being open in that area long enough to settle to get that accuracy. Puck movement for a one timer, as long as the lane was open doesn't give the defense as much time to disrupt that shot.
In most cases with one timers, getting the puck into the top half of the net where the goalie isn't will yield a goal but you are right, the less the goalie has to move over, the higher his chance is to cover more net as he doesn't need to stretch out as much like he would trying to take away the lower half of the far side.
How much the goalie commits to the original puck carrier also weighs as a big factor in the goalies ability to read and react quickly to the pass or not. If that puck carrier is skating in with his puck to his backhand curling away from the net as a non threat to score, the goalie is more likely to be anticipating pass and will get over much quicker than if the player is threat to shoot and you get the goalie to commit with some momentum before making the pass against it.
Best way to score, besides just walking in and exploiting the butterfly goalie animations is to be wide for sharp angle shots. There's sweet spots. Again it's about learning the animations and exploiting them. Thats why 3v3 is simply about ragging the puck to draw penalties so you can get penalty shots and do bh/fh/bh all day for 100% goal scoring.
I'm gonna look at more video to investigate some stuff. Because I feel that their reaction is maybe too high from one timers? I'm not sure if that's something that can be modified with your system or not but I'm going to look at and see if the shots from the middle slot are being gloved/blockered as opposed to bodied.
They will sometimes be saved with reactionary saves with their limbs as well. I am just saying that overall, their reaction time and precision will be slower the more they have to move and the less settled they are. It isn't black and white that you will never see it. Goalies make great glove saves on one timers in the real world too but those are more incredible saves than glove saves on longer weaker shots they are able to see easily.
On one timers, goalies will blindly/pre-emptively slide over and take away as much net as they can after they read the pass. On a wrist shot, that is essentially getting big in butterfly and the better the angle of the shot, the more net you have to shoot at. A goalie tries to get their positioning right to take away as much net as possible. On one timers, the side of the net they need to get to is more obvious so they can commit with their leg and arm to that side but that is why they also get beat against the grain or easily up and over their leg/arm into a bigger area of the net than is usually there with a good angle cut off on a regular wrist shot -- but in either case, a great shot can beat the goalie if they can't react in time which is why accuracy is a big piece to the puzzle as well.
I understand what you're saying, but being that close to the net, there shouldn't be enough time to react unless they're just getting a piece of it really. Most goalies throw themselves into position on one timers and hope the puck hits them, react secondary. I'm just saying often the goalies are reacting too quickly for the one timers, but not reacting enough sometimes for shots at distances.
I believe you saw that video with the tip in front where the goalie who has no chance somehow reacts with the glove. Can't remember which thread it's in. You see those, and it almost feels as if the game is pre programmed to make saves. It's likely not that way, but if you saw those saves against you and your goalie is allowing clappers far side from 5 feet off the top of the circle you could use your imagination and almost feel as if the game is fixed.
Right, but my point is that the two things aren't even in the same ballpark. Even if it's a 95-mph shot, if it's an unscreened shot taken from 55 feet away, that gives the goalie .40 seconds to react and make the save. That's more than enough time for an NHL goalie to make a save, especially given that he has a clear view of the shooter and so can read the puck right off the stick.
With a deflection off a slapshot, there is essentially no time to make a reaction save. If the guy (as in the clip below) is 16-18 feet away (let alone closer), even with the deflection slowing the puck the goalie's going to have .20 seconds or less to react. That's close to the minimum possible reaction time for a human (even a well-trained human) to even start to move. And the goalie can't anticipate, because by definition a deflection can go anywhere. So if goalies' reactions in this game are good enough to make a deliberate save like this, where the goalie tracks the deflection and reaches up to knock the puck away (rather than just making himself big and getting lucky):
they should basically never be beaten from distance on unscreened shots, unless you've coded the game so goalies are in effect allowed to cheat on deflections (by giving them a heads-up about where the puck is going).
This is basically the same thing HoodHoppers just said in his post: goalies' reaction times on deflections and close-in one-timers seems totally different (as in, much faster) than their reaction times to above-the-circle shots.
You actually responded to my post that explains why this is possible but I should have explained further. The goalies don't react the exact same on every single shot. There is human error built in which is what their attributes factor in as well as other factors in terms of what is happening in front of them. They are constantly analyzing if they think the player is going to shoot, pass or skate/deke with the puck, they are recognizing other threats around them and physical factors such as their movement/momentum and time to move their body/limb to the puck play into it as well.
Again, I am glad that you look into things that detailed. Before I responded to you the first time I had actually broken down similar scenarios to have some math done on it as well regarding a deflection coming in at 45mph from 18-22 feet out vs a wrist shot of 70ish mph from the top of the circles vs a slapshot of 90mph from 55 feet out similar to what you did and it really comes down to a difference between .3 and .4 seconds. I don't think you are compensating for enough speed being taken off when a puck is deflected by as much in that example but I don't think that is the main thing to worry about as we are talking relatively similar numbers and it is all happening very quick with little room for error on either side. I had the luxury to try a few deflections in our test bed where I can look at the speeds to see as well and feel 45mph is actually generous. (I couldn't remember if I posted anything about this or had just thought about it so apologies if it repeats anything)
The difference of 1/10 of a second more delay before he starts to try and make his save or not can allow the slapshot to score in case with the same save reaction that he responds to right away on a deflection with that speed/distance. That also assumes the puck location relative to the goalie is the same as well. A deflection in our game would have more accuracy to the perimeter of the net more often than a deflection would.
In most cases the goalie will have more delay and won't be reacting right on point so shots that get to him quicker have a higher chance to score in all cases than those that don't when all is the same but at his best, he can still react quickly with a good read and it just depends how long it takes to get his limb to the puck to get a piece of it. A bit more delay in the deflection case and it's in the net and you can take plenty of slapshots from the blueline and have the goalie save them all day. So overall, it aligns with what you are saying.
And you are right, for a perfect simulation, maybe our min reaction times are too low, maybe our max delay is too high, maybe our max shot accuracy is too high but we have always said that on competitive game style everything is a bit heightened for 4 minute periods and to feel balanced and not random. It still leans quite a bit to the sim side as far as our game styles are concerned though.