EA Forums - Banner

An analysis of the Current Rock State Meta - Geoff Harrower aka GamePlayDevUFC

EA_Roger
1445 posts Community Manager
Principal Software Engineer, Gameplay & Animation developper, Geoff Harrower took the time to give an analysis of the rock state meta in UFC 3:

There's been some discussion about the rock meta on Twitter, and I wanted to take the opportunity to present some analysis I did on a few scenarios that may give some high level players a few options to try out.

My motivation in doing this analysis was to disprove the argument that the rock state is a 25/25/25/25 dice roll.

The first rebuttal is an obvious one. If you are rocked with a relatively healthy chin and/or max health, or you were rocked with a relatively low stun value, the rock state will not last very long and you are probably in a position where you don't need to use any defensive strategies to survive. You can likely just turtle up and eat the block break and still survive.

Deeper into a fight when you've taken some damage, this may not hold true. There is certainly a skill element in reading the scenario, and deciding whether or not you need to take the chance of using head movement to survive, at the risk of getting caught and having the fight end.

Now, lets assume you're in a position where a block break would end the fight. What options are available to you?

The option most people are aware of is to read the combo your opponent is throwing, and anticipate a particular strike arc and use head movement to avoid it.

In all but the most dire of situations, this will likely mean reading a four strike combo and making use of head movement on the second, third or fourth strike before the block breaks.

The advantage of waiting to the fourth strike is that you've bought yourself time to recover before making your move, so that if you guess right you are more likely to survive.

The disadvantage is that you will likely eat damage due to partial block bleed through along the way to the fourth strike.

Again, reading the situation your are in will be critical to survival.

But what if you don't want to rely on guessing and reading your opponent's patterns? What if your opponent is really good at mixing up block break combos and there is nothing to read?

Are you really just left with a guessing game?

My argument is no. There is an option that removes guess work and relies on fast reactions. It's by no means easy, but I believe it to be theoretically viable.

To understand it, lets look at damage breakdown in several scenarios.

Using a McGregor mirror match, let's look at the damage done using a lead head kick from punch range.
RYAnqKH.gif

As you can see, this particular kick from this particular range does 20 damage.

Now let's look at the damage done when you sway back as the kick lands.
aphfKga.gif

In this case, swaying back increases the range, making the kick deal 30 damage.

Now, what happens if you back sway and then chain to a side sway away from the kick at the last second?
Arepi59.gif

In this case, the kick only does 9 damage.

That's more than a 60% reduction in damage. When your goal is survival, that's a pretty good outcome.

Let's try another scenario. What if you sway to one side, then sway to the opposite side just before the kick lands?
qDkfs4O.gif

In this case, the kick does damage of 13.

So what does this have to do with the rock state?

What this allows you to do, is sway mid combo in anticipation of the most likely strike to come out next, and then re-actively sway away if you're wrong and something else comes out.

Of course this approach would only be successful if you have enough time to react to the animation and sway in the correct direction.

Fortunately, the heaviest strikes in the game are also the longest.

Head kicks and overhand punches mid combo after a successful block all give more than a 22 frame window to enter the sway and receive damage mitigation.

Although fast, this is within the range considered reactable based on tuning done for the grappling game in UFC 2. Your mileage may vary once online latency is thrown into the mix.

So with the damage properties understood, and if you are willing to accept that this technique falls within reaction time thresholds, the next important part is understanding the combo tree.

If you back sway the third strike, you will avoid hooks and uppercuts no matter what. If you are in kick range you will avoid straight punches but not kicks. If you are in punch range you'll get hit by both head kicks and forward moving straight punches. And straight punches are too fast to react to.

The back sway carries with it vulnerability to straight punches, so that's important to be aware of. Getting caught with a jab while back swaying does 10 damage, so not much worse than a mitigated head kick.

But a straight punch will do almost the same amount as a block broken head kick, so you cannot afford to get hit with that.

So the back sway to side sway strategy is only appropriate in situations where you know you won't get hit by a straight on the next strike. Either because of range, or because that last strike thrown in the combo was with the back hand.

By contrast, the side sway strategy can be used if you know the only strikes that will come out next come from the side you are swaying away from OR a strike that is slow enough to react to.

The one branch of the combo tree that really messes with this strategy is the jab to lead/back hook.

Once a jab has been successfully blocked, neither the back sway to side sway or side sway to side sway strategy will keep you safe from a jab to rear hook or jab to lead hook mix up. You always face a 50/50 chance of swaying into a hook if your opponent chooses that particular mix up.

Without fully exploring all options in the combo tree, I feel it's safe to assume that there is a set of combos that could force your opponent into this dilema after a fair bit of block breakdown and/or bleedthrough before reaching the fully broken state.

Given the current state of the meta, I feel the strategies above are viable but must be executed before a full block break on a strike that is not a jab.

The viability of the above strategies also degrade as the fighters chin, long term health and long term block are reduced later in a fight.

In my analysis, I was hoping to find a viable survival option through any branch of the combo tree that did not rely on guess work, but this dilemma with the jab to lead/back hook mixup prevented that.

I may have some updated analysis in the future to help people deal with that issue as I feel it's really the only scenario that partially validates the complaints that the rock meta is a dice roll.

Stay tuned.

Until then, I hope the information provided here gives people some new strategies to try out in their efforts to survive the rocked state.

Follow @GeoffHarrower on twitter for more updates.
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!