Backup Endurance (Exploit/Test)

[larrymcg421]

In the Q&A thread, I got the impression that endurance was only measured for backups and not for starters. To test this theory, I started a new career and picked up a Running Back (Bo McDaniel) with an endurance rating of 0/25 (all green), which was the lowest I could find.

I ran two tests. In the Normal Test, I simply put McDaniel in as the starter and put his playing time at 100%. In the Flipped Test, I put McDaniel in as the backup but made RB2 as the starter in every formation setting. The true starter was still listed with 100% playing time. I ran 5 tests under each setting (I could have run more but there was very little deviation)

Here are the results:

<table border="0" cellspacing="0" cols="3" frame="void" rules="none"> <colgroup><col width="86"><col width="86"><col width="86"></colgroup> <tbody> <tr> <td align="left" height="17" width="86">
</td> <td align="left" width="86">Attempts</td> <td align="left" width="86">Yards Per Carry</td> </tr> <tr> <td align="left" height="17">Normal </td> <td sdval="153" sdnum="1033;" align="right">153</td> <td sdval="2.95" sdnum="1033;" align="right">2.95</td> </tr> <tr> <td align="left" height="17">Flipped</td> <td sdval="351.6" sdnum="1033;" align="right">351.6</td> <td sdval="3.48" sdnum="1033;" align="right">3.48</td> </tr> </tbody> </table>

A few things are obvious from this test...

Not only can we see that endurance is not factored in for backups, but there's a dramatic increase in Yards Per Carry under the Flipped scenario eventhough he's getting twice as many carries.

Jim mentioned that commissioners may want to block exploits like this, so this is a heads up. One thing thats problematic is that its not obvious to detect. McDaniel was still listed as starting between 14-16 games each season on the stat sheet.

I'll do some more tests later to see how this effects players of differing abilities (McDaniel is a subpar RB).

[Ben E Lou]

Larry, I hadn't thought about doing it this way. This is an even larger exploit than has been known. I just ran a duplicate test. It's not so much how many more carries; it's that the guy slotted as RB2 in this scenario gets 100% of his team's carries. So, yeah, that's a clear exploit, no question about it. The RB2 with 4/8 endurance I just tested using your idea here got 543 carries for 2233 yards the first shot using your method (out of 543 RB carries on the team). Slotting him as RB1 with 100% endurance and straightforward formations, he got 292 out of 546 RB carries. Just to be certain, I ran two more seasons with the RB2 slotted as the #1 back: he got 100% of the RB carries each time--over 500 carries.

The good news is that there is an obvious "tell" for this: if your opponent has a guy listed as RB1 in the depth chart, but he gets no/few carries, it's clear that he's doing something fishy.

[MIJB#19]

I noticed something similar in testing when I used several combos of running backs in what started as an attempt to figure out how to just get my 3rd down back to get 3rd down carries. I didn't use this RB2 starts in a formation thing, but I did use the 3rd down back RB slot for 'obvious' short yardage formations and the passing downs RB slot for 'obvious' passing formations. That still left a lot of formations where RB1 was the normal starter. After some testing I noticed that if my 3rd down and passing downs specialist was also my RB2, that player would get almost 100% of the 50ish 3rd down carries (96% to be precise), but sadly the side effect of this was that in the non-3rd down carries, RB1 got about 150 carries and specialist-RB2 got about 200. I tried it with a high endurance starter and a medium endurance starter and the carries distribution between the two RBs was roughly the same.

I had mixed feelings as I did get to a point where I could get my 3rd down back to get 96% of the 3rd downs carries, but it also meant that my true RB1 was getting only 39% of the total carries and RB2 61%. I tested it with a high endurance RB and a medium endurance RB and the percentages where amazingly identical. Note that QB carries were ignored in testing and the playing time was set to 100% for RB.

I also noticed that when using the RB1 as my 3rd down and passing downs specialist would lead to the most realistic carries distribution based on starter endurance, a very high endurance guy (86) got 75% of the carries, a medium endurance (43) guy got 68%. The downside concerning my initial intent was that my all-downs RB was giving up 3rd down carries to RB2, with the high endurance RB getting 79% and the low endurance RB getting 68%.

When I re-did that all-downs RB test with a 50% starter setting, the carries distribution went into the 50-50 neighbourhood, yet the 3rd down carries fluctuated from 51% to 64%.

FWIW, unlike running multiple seasons, instead I ran full season tests against 16 identical superior opponents. Although a smaller sample size, I figured this would at least tone down on the opponent specific success and decrease the randomness a bit. (About that randomness, I think I ran about 50 test seasons in that environment and always saw my team end up with 8-8 or worse record in all but one try (and that was a 9-7 record), however, opponents could have their star RBs go anywhere between 5.2 and 3.5 ypc, have their star QB rack up a passer rating of anything between 120 and 80).

[Icy]

What I was thinking and this proves it, is that the engine checks the starting RB endurance, and if the engine thinks that the starter is tired, then the backup gets the carries.

That is fine but...it seems the the endurance/tiredness check is not applied to the backup, so the backup endurance doesn't matter at all, he will get all the carries that the starter RB can't run like if he had infinite endurance. Also if you put RB2 as starter in every formation, he will get 100% of the carries as his endurance is never checked.

The easy fix would be to also check the endurance/tiredness for the backup RB, in fact i don't know why it's not being done right now.

The system should work this way:

- If RB1 is too tired, RB2 should replace him
- If RB2 is too tired, it means RB1 had time to rest already, so he should get back in the game, and even if RB1 is not totally recovered, he should be back in the game, as we all would prefer to play a tired RB1 than a tired RB2.
- Endurance should be tied to performance decrease with playing time and not directly to playing time. Else i see no point on the % we set in the deep chart. A 100% should mean that the RB1 gets 100% of the carries, a 50% should mean we want the two RB's to equally share the load, etc, and are we who should determine if we want our RB to carry the ball 100% of the times, looking at his endurance and the possible performance drop when tired and setting the adequate % in the deep chart.

[cuervo72]

Larry, I hadn't thought about doing it this way.

Seriously? I mean, you really hadn't? I figured doing something like that - at least *trying* - it was obvious. Especially when jkat came out and suggested that you could fiddle with things put your best WR in a slot where he couldn't be double teamed, I figured there were other ways of possibly tinkering with this.

And I know I've mentioned it before, but there's also a loophole on the defensive side, if you have low endurance starters. Just run a whole lot of nickel and dime, as the slotted starters there don't seem to be substituted for. It might not mean great things overall for your defense (though it may not hurt all that badly), but it seems like a method for forcing your starters to be in there.

[Ben E Lou]

Seriously? I mean, you really hadn't? I figured doing something like that - at least *trying* - it was obvious. Especially when jkat came out and suggested that you could fiddle with things put your best WR in a slot where he couldn't be double teamed, I figured there were other ways of possibly tinkering with this.Well, the way I had always looked at the issue was "If RB1 endurancecountdown < x, then use RB2." I never really thought about the impact of that on the usage screens, though. Yeah, looking back, it should have been obvious. I can't believe I never thought about it with my high-endurance (and therefore low-carry) RB in IHOF FOF2K4. (I checked it this morning, and sure enough, this would have been a workaround for that issue in FOF2K4).

And I know I've mentioned it before, but there's also a loophole on the defensive side, if you have low endurance starters. Just run a whole lot of nickel and dime, as the slotted starters there don't seem to be substituted for. It might not mean great things overall for your defense (though it may not hurt all that badly), but it seems like a method for forcing your starters to be in there.I brought this up in a different arena in IHOF. Across the board, the game doesn't reach outside of the depth chart at a given position group unless there's an injury in that group. And since the nickel/dime depth charts have no backups, those players will play 100% of the time in those situations, barring injury.

[highfiveoh]

What better? Endurance in OOTP or FOF?

[larrymcg421]

Skydog, this wasn't exactly what I was talking about in the other thread, just something I came across while testing if backup endurance is calculated. I agree that this is clearly an exploit.

I was simply going to flip my depth chart (but not the RB1/RB2) to get a more realistic distribution of carries. It seems like that would be considered an exploit, but if so, then I'm already exploiting the system as is, because my low endurance backup is getting way more carries than he should. It sounds like we're saying it's okay to exploit with your 2nd best RB, just not your best.

Also, that leads to the problem of determining when an owner is exploiting. If an owner has two low endurance RB's, how do we decide which RB the owner has to start?

[Ben E Lou]

Larry:

I agree that it's a murky discussion. I think this is a "when in doubt, check it out" situation--let the league administration know ahead of time how you intend to set the depth chart and playing times for him. I think the key issue with that back in particular is that he's just got very low endurance and therefore shouldn't get many carries at all.

[Icy]

I guess it's the same also for any position, if you set your best TE as TE2, btu then you place TE2 as the TE in every formation, he won't ever be replaced either.

[Vinatieri for Prez]

Too late for a Jim fix on this one unfortunately.

[Ben E Lou]

This appears to have been addressed in 6.1a. Some testing has revealed that low-endurance backups are performing worse when they're left in for the entire game. Good news here!

[wade moore]

This appears to have been addressed in 6.1a. Some testing has revealed that low-endurance backups are performing worse when they're left in for the entire game. Good news here!Can you elaborate at all? Is this only when using this exploit, or is a low endurance guy set at 100% (with say a high end guy at RB2) going to have problems as well?

[Ben E Lou]

Can you elaborate at all? Is this only when using this exploit, or is a low endurance guy set at 100% (with say a high end guy at RB2) going to have problems as well?I didn't specifically test for the latter, but the "sniff test" says that's not the case, because I did set the low-endurance back to 100% playing time for my baseline test. Here are the 10-season comparisons. Leggin is an all-around stud with 83 endurance. Edwards has 80 hole recognition and some other solid running bars, but has 4 endurance.

EDWARDS (LOW-ENDURANCE) WITH ALL CARRIES: 3.58ypc
EDWARDS AS #1 WITH 100% PLAYING TIME: 4.37ypc

I didn't record the carries for Edwards vs. Leggin each year for the latter test, but I just ran another one, and the 245 vs. 195 split that I just got looks similar to what I was getting.

LEGGIN (HIGH-ENDURANCE) WITH ALL CARRIES: 4.26ypc
LEGGIN AS #1 WITH 100% PLAYING TIME: 4.65ypc

The one-season look with Leggin as the #1 back had him with 395 of 482 RB carries, or around 82%, fwiw. Given how backup RBs tend to perform in FOF, I'm not even convinced that you're better off giving your stud the rock on every carry now, unless you have a *really* bad backup.

[wade moore]

:thumbsup: - like you said, at a sniff it looks like putting a low-endurance guy at 100% isn't a problem as long as there is a backup that can spell carries for him.

[Ben E Lou]

:thumbsup: - like you said, at a sniff it looks like putting a low-endurance guy at 100% isn't a problem as long as there is a backup that can spell carries for him.
I'll have to go back and check the Q&As. I believe Jim said that in either 6.1 or 6.1a, you can get into trouble if both of your RBs are low-endurance.

[wade moore]

I'll have to go back and check the Q&As. I believe Jim said that in either 6.1 or 6.1a, you can get into trouble if both of your RBs are low-endurance.
This would seem to show why as one way or another you've got a RB getting more carries than they can handle.

[Ben E Lou]

Never mind that. He said it's something he'd like to add down the road, not in a patch. Maybe whatever I'm seeing is related to the bolded part below.

Further tweak running back choice algorithm. Study real-life limits, remove part of the exploit for putting low-endurance backup on chart.

[wade moore]

Never mind that. He said it's something he'd like to add down the road, not in a patch. Maybe whatever I'm seeing is related to the bolded part below.

May be worth trying the "put him in all of the formation subs" exploit to see if it has the same effect?

[Ben E Lou]

Uh, Wade, that's what we're talking about here.

[Subby]

If I had a nickel for every sentence that started.. "Uh, Wade..."

:D

[zbuckley]

\\ Especially when jkat came out and suggested that you could fiddle with things put your best WR in a slot where he couldn't be double teamed, I figured there were other ways of possibly tinkering with this.


I've thought about putting my best WR in the slot due to his skill set not trying to avoid double teams. He has 100 Getting Down Field and 85+ RR with only 45 BPR. Seems ideal for the slot but is that cheating?