Question to KEM farmers

"sflowers330;1592069" wrote:
For clarification, the cost of ensuring 3 bonuts is $50k on average.

Note, this assumes that the probability of the 3 donut choice is equally spread across the boxes, which in my experience is not the case.

I know this is technically correct, but it's a little misleading for someone trying to optimize their output (and I know you hinted at that in an earlier post, I'm just showing the math for the newcomers to farming)

In this kind of stat, you don't want to include the average when you get 3 donuts with your first pick because there is no cost and no choice to make, you've already got the optimal result. The question is really "what do I do if I only get 1 or 2 donuts?" in which case the real cost of getting to 3 donuts is $75K on average.

This means if you already got 2 donuts, it costs an average of $75K to get just one more donut. And if you only got 1 donut, that average drops to $37.5K per extra donut.

The higher your % goes, it becomes more cost-efficient to spend that money on KEMs, because a new bonut box will get you an average of 2 donuts on your first pick. And don't forget, you get 25% of the KEM purchase price back, so that break-even point comes earlier than you expect.

So give or take a few % points, at about 750% multiplier (including your hidden SH bonus) you should stop picking a new box if you get 2 donuts with your first pick.

At about 1200%, if you get one donut you should open a second box and keep whatever you get.

And at about 1500% total multiplier you should stop picking a new box even if you get 1 donut.


Now keep in mind I'm ignoring external factors like the time and space it takes to have a large KEM farm, and I'm only showing the numbers for farming a significant # of KEM at a time where the cost of a KEM on average is close to the max $14,400.

If you are only farming 20 KEM at a time where the average cost of each KEM is WAY cheaper, you should stop paying for extra picks at about half the bonus % shown above.

Happy to answer questions about the math, just wanted to give some good 'rule of thumb' calculations for those interested in really cranking up their farming.

@moviefansam I'm not sure I agree with your statement re ignoring the first choice. On these stages you need all three donuts and no choice still factors into the number.

If you're trying to optimize your cash flow you need to understand what the average cost will be per bonus, so you need to include all the possible outcomes. 50k is the correct number to use. If you're intentionally overstating the costs to stay conservative that's one thing, but the results will be suboptimal as you could have made more KEM​.

"sflowers330;1592143" wrote:
@moviefansam I'm not sure I agree with your statement re ignoring the first choice. On these stages you need all three donuts and no choice still factors into the number.

If you're trying to optimize your cash flow you need to understand what the average cost will be per bonus, so you need to include all the possible outcomes. 50k is the correct number to use. If you're intentionally overstating the costs to stay conservative that's one thing, but the results will be suboptimal as you could have made more KEM​.

Hmm. Sorry, I agree with @moviefansam ’s logic on this one.

i'm not taking credit for this, as some kind soul posted it from here. i'm verysorry i can't remember who it was

http://lwshost.com/tsto/simkem.pl

you can see there what you should do for hte number of KEM you make, and the bonnus % etc.

"jellypawn;1592173" wrote:
i'm not taking credit for this, as some kind soul posted it from here. i'm verysorry i can't remember who it was

http://lwshost.com/tsto/simkem.pl

you can see there what you should do for hte number of KEM you make, and the bonnus % etc.

Thanks, that's pretty cool. I should never open a box, but I can't help it!

"docmorrissey;1592186" wrote:

"jellypawn;1592173" wrote:
i'm not taking credit for this, as some kind soul posted it from here. i'm verysorry i can't remember who it was

http://lwshost.com/tsto/simkem.pl

you can see there what you should do for hte number of KEM you make, and the bonnus % etc.

Thanks, that's pretty cool. I should never open a box, but I can't help it!

Don't feel bad... play however you prefer to play! :)

I found it interesting to experiment with that simulator - it's useful to see the effects of changing strategy, but in some ways "best strategy" is also quite tricky to define. Trying to minimise the average $ spent per donut is great, however always involves a trade-off against the effort involved, so we might all have different points at which we just shrug, say something like "meh, it's not worth it" and make our own choices!

When I plug my game statistics into the simulator, it suggests that I shouldn't be going for 3 donuts from every round of bonuts anymore... however, if I choose the more efficient strategy then of course I receive fewer donuts for each iteration, so I have to build more KEMs each time to make up the difference. I can feed those increased numbers back into the simulator until the yield is again pretty similar to my original strategy... it then reports that two of the alternative strategies become very close to most efficient; some simulation runs produce different winners because there's so little distance between them and overall I may need to build as many as 30% more KEMs for a pretty marginal increase in efficiency, as well as maybe take a little more momentary thinking time to apply a different strategy when tapping the boxes. So potentially a significant amount more effort for only a little gain. On that basis, I'm still fairly happy harvesting slightly fewer donuts to save the extra time and enjoy the game more (although I'll continue to review my strategy at higher multiplier levels)!