Warrior General Games

Originally posted by Larry Roadgrader
That one Bort even chimed in on and clarified that 11 (or more) is greater (in effect) than 10. Sure, there's almost certainly a diminishing return, but to what degree we don't know, and to pretend that there is universally no use to going over 10 is silly.

I'm sure it's similar to the AEQ percent pieces. Going over 11 is probably like adding a 3rd piece of % gear, and going over 14-15 is probably like adding a 4th.

Yeah. Diminishing returns is a really interesting concept. If each point gives 90% of the previous, then we're looking at:

10 = 10
11 = 10.9
12 = 11.71
13 = 12.44 etc

But if Tpaterniti is correct about it halving every time, then you couldn't even reach a value of 11 with a SA.

10 = 10
11 = 10.5
12 = 10.75
13 = 10.875 etc

That's also assuming an exponentially diminishing return. It could be that each point over 10 just loses a 10th or something.

10 = 10
11 = 10.9
12 = 11.7
13 = 12.4 etc

Or that every point over 10 is just worth a flat amount (though I think he debunked this one). So, using .75 it would be.

10 = 10
11 = 10.75
12 = 11.5
13 = 12.25 etc

Point being, telling us it's a diminishing return doesn't actually tell us anything useful at all.

Originally posted by Go Pack 120
Point being, telling us it's a diminishing return doesn't actually tell us anything useful at all.

Well, it is useful to know it's a diminishing return vs. not know it's a diminishing return.

It's fun to hypothesize and tweak to see the effects, but I think the code is about 1/10th as complex as most of you give it credit for. Bort and his team clearly did a great job keeping us old nerds interested/intrigued for so long, but this isn't Madden developed by EA Sports with a $60 million annual development budget and a 200 developer team.

It doesn't have to be a complex system to give complex results though. And RNG just adds a flavor of "Hell if I know" to the whole situation.

Fwiw I agree with the packer fan (the horror!).

Also I think the correct historical knowledge is that diminishing returns are at greater than 11. 11 existed from the origin of the game, and diminishing returns on SA's weren't discussed or considered until AE SA bonuses. Even the vague bort quote everyone does shit at interpreting, don't state 11 in an SA is diminishing.

It's worth noting that I've seen quite a few "special" WL dots with 14-19 in a given key SA. Especially if it's an exponential SA where each level increases both chance to fire and power when firing.

Originally posted by WiSeIVIaN
Fwiw I agree with the packer fan (the horror!).

Also I think the correct historical knowledge is that diminishing returns are at greater than 11. 11 existed from the origin of the game, and diminishing returns on SA's weren't discussed or considered until AE SA bonuses. Even the vague bort quote everyone does shit at interpreting, don't state 11 in an SA is diminishing.

It's worth noting that I've seen quite a few "special" WL dots with 14-19 in a given key SA. Especially if it's an exponential SA where each level increases both chance to fire and power when firing.


Great post! Love the "11 is the baseline assertion"!

Maybe the diminishing returns could be the term used for one questioning the SP cost of the SA vs what other things that you could do with those points? Once the cost to apply those gets to a certain range it becomes less efficient?

Originally posted by Go Pack 120
Yeah. Diminishing returns is a really interesting concept. If each point gives 90% of the previous, then we're looking at:

10 = 10
11 = 10.9
12 = 11.71
13 = 12.44 etc

Pretty sure it has to be stronger diminish than this. My brother built a couple pHBs back in the ~50ish seasons, with 22.5% BT, and 18 Power Thru. And they were straight up the worst pHBs he's ever made for my squads.

So if we use that example,
10.9, 11.71, 12.44, 13.1, 13.69, 14.22, 14.7, 15.1 is where that would've arrived at.

If it is just a straight repeated % diminish like that, I'd guess something more like ~75-80%? Or hell, maybe it is 50% like TPat suggested? All I know is, I don't think either of us have built anything above a 13ish SA since, where it seemed like there's a punishing effect going loopy on it, like an 18. Or maybe even Bort expected people would try to abuse PThru, and made THAT specific one have more diminish to it? Just spit balling, I haven't actually gone through and collected fire rates on similar dots, so this is all just my speculation, ofc.

Originally posted by Gambler75
Pretty sure it has to be stronger diminish than this. My brother built a couple pHBs back in the ~50ish seasons, with 22.5% BT, and 18 Power Thru. And they were straight up the worst pHBs he's ever made for my squads.

So if we use that example,
10.9, 11.71, 12.44, 13.1, 13.69, 14.22, 14.7, 15.1 is where that would've arrived at.

If it is just a straight repeated % diminish like that, I'd guess something more like ~75-80%? Or hell, maybe it is 50% like TPat suggested? All I know is, I don't think either of us have built anything above a 13ish SA since, where it seemed like there's a punishing effect going loopy on it, like an 18. Or maybe even Bort expected people would try to abuse PThru, and made THAT specific one have more diminish to it? Just spit balling, I haven't actually gone through and collected fire rates on similar dots, so this is all just my speculation, ofc.

Anecdotal but this dot from grandmaster Shink had a good bit of Power Thru and led CL in rushing a bunch of times, despite a horrible OC: https://glb.warriorgeneral.com/game/player_awards.pl?player_id=4848191

Keep in mind something like Power Thru or first step, each point increases both the chance to fire, and separately the strength when firing. This means if any SA's are worth taking past 11, these type of SA's would be the likely ones to fit.

Originally posted by WiSeIVIaN


Keep in mind something like Power Thru or first step, each point increases both the chance to fire, and separately the strength when firing. This means if any SA's are worth taking past 11, these type of SA's would be the likely ones to fit.

Yes, I've always remembered you pointing this out--great point.

Originally posted by Go Pack 120
Yeah. Diminishing returns is a really interesting concept. If each point gives 90% of the previous, then we're looking at:

10 = 10
11 = 10.9
12 = 11.71
13 = 12.44 etc

But if Tpaterniti is correct about it halving every time, then you couldn't even reach a value of 11 with a SA.

10 = 10
11 = 10.5
12 = 10.75
13 = 10.875 etc

etc

Point being, telling us it's a diminishing return doesn't actually tell us anything useful at all.

I don't think looking at it as % of an SA add is the best way to think about it. Most of these add some bonus to a roll or situation, so probably that is a better way of looking at it. 16 of Shed Block is really 16. It's just that the bonus to the roll at that point (adding from 15 to 16) may be +0.25% whereas it may be 2-3% at a lower level. But 0.25% may still be significant. You may have to add from 15-18 to add 1% or 0.5% of a bonus whereas adding from 10 to 11 may give you a 1-2% bonus to the roll.

We also don't know if the bonus is constant from 1 to 10 or if it diminishes from there. Another complicating factor is that many of the outcomes are a plateau in the sense that you either achieve it or you don't .You either trigger a diving tackle or you don't. There is also a RNG in the mix, so the best way to look at this may be more like 6 of Diving Tackle is enough to trigger the SA 60% of the time a player attempts a tackle, whereas 11 is enough to trigger it 85% of the time, and 15 is enough to trigger it 88% of the time. This also depends on your attributes. If that player has 80 Jumping and 90 Tackling, then maybe 6 is enough to trigger it 75% of the time, 11 is enough to trigger it 90% of the time, and 15 is enough to triger it 92% of the time. At least this is how I think of it.

By the way, the Strategy Expert's building guide was all about getting the most for your money, which has turned out to be a completely wrong way to approach this game. More extreme builds designed to beat certain rolls even if they are not good all around seem to be way better than good all around builds. I have built all of these, including things like a team full of players with 60 confidence each. It makes the overall bars look good, but they don't play that much better.