Endurance Case Study

[sabotai]

A question as to what exactly does Endurance mean came up on the eNFL boards. So, I decided to run a little case study.

I take 2 players, 1 from a human controled team (Strahan, NYG) and one from a computer controled team (Sapp, TB).

I do this so that I can control the playing time of the human control player, to see the effect of the stats, and so I can see how the CPU team handles players with high/low endurance.

I'll create 5 games with both Sapp and Strahan having a high endurance rating. Run it for one season and compare the stats (will turn injuries off)

I'll then run it another 5 games with it at high with injuries set at 150.

I'll then repeat it with Sapp and Strahan having low endurance ratings.

Sound good? I'll post results once I get them.

[sabotai]

Little tidbit I ran across already. When injuries are set to 0, injuries like Food Poisoning still occur.

[sabotai]

Games 1-5: Strahan/Sapp high endurance, Injuries = 0

Game 1

Strahan Endurance = 88
Sapp Endurance = 93

Strahan - RPY = 425, PPL = 545, 53 tk, 13 ast, 11.5 sack, 7 blk, 35 hur, 9.8 PR%, 7.9TK%

Sapp - RPY = 435, PPL = 534, 61 tk, 32 ast, 2.0 sack, 3 Blk , 10 Hur, 2.8 PR%, 9.8TK%

Game 2

Strahan = 85/92 Endurance
Sapp = 96/98 Endurance

Strahan - RPY = 455, PPL = 500, 69 tk, 15 ast, 13.5 sack, 3 blk, 31 hur, 9.5 PR%, 10.2 tk%
Sapp - RPY = 422, PPL = 547, 77 tk, 29 ast, 8.5 sack, 2 blk, 19 hur, 5.3 PR%, 11.8 TK%

Game 3

Strahan - 98 Endurance
Sapp - 98 Endurance

Strahan - RPY = 476, PPL = 564, 62 tk, 21 ast, 17 sack, 9 blk, 30 hur, 9.9 PR%, 9.6 tk%
Sapp - RPY = 425, PPL = 566, 63 tk, 31 ast, 6.5 sack, 1 blk, 17 hur, 4.3 PR%, 10.1 tk%

Game 4

Strahan - 85 Endurance
Sapp - 88/95 Endurance

Strahan - RPY = 425, PPL = 548, 48 tk, 19 ast, 11.5 sack, 0 blk, 26 hur, 6.8 PR%, 8.0 tk%
Sapp - RPY = 381, PPL = 541, 52 tk, 21 ast, 9 sack, 2 blk, 23 hur, 6.2 PR%, 8.8 tk%

Game 5

Strahan - 86 Endurance
Sapp - 87 Endurance

Strahan - RPY = 490, PPL = 540, 60 tk, 16 ast, 13 sack, 6 blk, 23 hur, 7.7 PR%, 8.6 tk%
Sapp - RPY = 410, PPL = 553, 59 tk, 25 ast, 9.5 sack, 1 blk, 12 hur, 4.0 PR%, 9.7 tk%


Games 1-5 averages

Strahan - 454.2 RPY, 539.4 PPly, 58.4 tk, 16.8 ast, 13.3 sacks, 5 blk, 29 hur, 8.8 PR%, 8.9 TK%

Sapp - 414.6 RPY, 548.2 PPly, 62.4 tk, 27.6 ast, 7.1 sacks, 1.8 blk, 16.2 hur, 4.6 PR%, 10.1 TK%

Since I don't feel like managing rosters for 16 games 5 times over, both the Giants and TB will be cpu controled for the injury tests

[sabotai]

I'm axing the test for the injuries. After doing it, I'm certain I need a much larger sample size (lie doing it 30 times for each). Perhaps another day. Results of the low endurance stat tests coming up shortly.

[sabotai]

Games 6-10, Strahan/Sapp low endurance, Injuries set to 0

Game 6

Strahan - 8 Endurance
Sapp - 13 Endurance

Strahan - 465 RPly, 521 PPly, 45 tk, 21 ast, 5.5 sack, 5 blk, 20 hur, 5.8 PR%, 7.2 TK%
Sapp - 381 RPly, 485 PPly, 66 tk, 30 ast, 7.0 sack, 2 blk, 17 hur, 5.2 PR%, 11.8 TK%

Game 7

Strahan - 17 Endurance
Sapp - 14 Endurance

Strahan - 440 RPly, 508 PPly, 51 tk, 16 ast, 12.0 sack, 7 blk, 24 hur, 8.4 PR%, 8.3 TK%
Sapp - 380 RPly, 559 PPly, 60 tk, 20 ast, 11.0 sack, 5 blk, 11 hur, 4.8 PR%, 9.6 TK%

Game 8

Strahan - 2 Endurance
Sapp - 9 Endurance

Strahan - 449 RPly, 526 PPly, 44 tk, 15 ast, 12.5 sack, 6 blk, 36 hur, 10.3 PR%, 7.3 TK%
Sapp - 338 RPly, 498 PPly, 68 tk, 25 ast, 5.0 sack, 7 blk, 13 hur, 5.0 PR%, 11.7 TK%

Game 9

Strahan - 11 Endurance
Sapp - 0 Endurance

Strahan - 455 RPly, 539 PPly, 55 tk, 25 ast, 12.0 sack, 8 blk, 23 hur, 7.9 PR%, 9.2 TK%
Sapp - 339 RPly, 464 PPly, 44 tk, 17 ast, 4.0 sack, 2 blk, 19 hur, 5.3 PR%, 8.0 TK%

Game 10

Strahan - 9 Endurance
Sapp - 9 Endurance

Strahan - 429 RPly, 540 PPly, 55 tk, 20 ast, 12.5 sack, 2 blk, 33 hur, 8.7 PR%, 9.0 TK%
Sapp - 337 RPly, 467 PPly, 64 tk, 23 ast, 8.5 sack, 2 blk, 13 hur, 5.0 PR%, 11.8 TK%


Games 6-10 averages

Strahan - 447.6 RPly, 526.8 PPly, 50 tk, 19.4 ast, 10.9 sacks, 5.6 blk, 27.2 hur, 8.3 PR%, 8.2 TK%

Sapp - 355 RPly, 494.6 PPly, 60.4 tk, 23 ast, 7.1 sack, 3.6 blk, 14.6 hur, 5.1 PR%, 10.7 TK%

[sabotai]

Comparison

Warren Sapp

In his HE (High Endurance) seasons, he played an average of 414.6 RPly, but in his LE (Low Endurance) seasons, he only played an average of 355 RPly. A difference of 59.6 RPly per season. Similarly, in his HE seasons, he played an average of 548.2 PPly, and 494.6 PPly in his LE seasons. A difference of 53.6 PPly per season.

Since I have injuries set to 0, I don't beleive it is due to minor game injuries. I think it's clear that the CPU controlled teams will not play a low endurance player as much as a player with high endurance.

All of his stats are slightly lower in the LE seasons, but his percentages are slightly higher. Since I only did 5 seasons of each, they don't really appear to be statistically significant.

Interestingly, however, is that in his HE seasons, 1 out of 5 of them was a decently breakout year for him ("game 2") while in the LE seasons, 1 out of 5 of them was a decently "breakdown" year for him ("game 9").

It's interesting because Strahan suffered the same thing. 1 of his HE seasons was a minor breakout year (17 sacks in "game 3") while 1 of the 5 LE seasons was a decent sized "breakdown" year (5.5 sacks in "game 6").

Again, I'd probably have to repeat this study several times to see if these were flukes or a result of high/low endurance. But it's interesting nonethless.

[sabotai]

Comparison

Michael Strahan

I had him at 9 playing time, so his RPly and PPly numbers are abotu the same. Very slightly lower in the LE seasons, but not really significant.

Other than the interesting thing I pointed out in the Sapp comparison with regards to Strahan's numbers, I don't really find anything major.

He had more tackles in the HE season than the LE season (58.4 HE to 50 LE). Decent difference. But he had more assists in the LE seasons than the HE seasons (19.4 LE to 16.8 HE).

He had more sacks in the HE season than the LE seasons (13.3 HE to 10.9 LE). He also had fewer hurries in the LE season than the HE season (29 HE to 27.2 LE). Not really a big difference. More blocks in the LE seasons than the HE seasons (5.6 LE to 5 HE)

Keeping in mind that the Endurance difference between the HE and LE seasons was a large as possible, I find it hard to see (at least for Defensive Linemen) why Endurance would be a concearn when looking at players.

The only thing is if Endurance really does play a role in determining if the guy goes into a slump or has a big year then I can see looking at the rating. If you take out Strahan's slump in "game 6" and breakout in "game 3", the sacks are just about the same.

For now, it would seem that I'm going to side with the crowd that says they don't care about Endurance. Soon, I'll try the experiment with running backs to see what happens there as well. But again, these differences in Endurance were as large as possible. I can't see signing a player with 65 Endurance over a guy with 30 Endurance because of that difference (unless the 65 guy is exactly the same or better.) It just doesn't look like a big factor at all.

[amdaily]

I don't know where the thread is, but I once compared the attempts per quarter and yards per carry per quarter of RB's in an effort to see if endurance worked to wear down players performaces over the course of a game.

It didn't. Low endurance RB's got just as many carries and YPC in the 4th quarter as the 1st.

[MIJB#19]

I am still convinced players with endurance of below 50 are more vulnerable to little injuries. I have no large study proof of this, but it's too much coincidance that on my IHOF team, most if not all injuries happened to players with endurance below 50. Not counting the QB injuries as their endurance is hidden.

[Subby]

Thanks for doing this study - it largely confirms my suspicions about the effect of endurance in the multi-player arena...

[sabotai]

I am still convinced players with endurance of below 50 are more vulnerable to little injuries. I have no large study proof of this, but it's too much coincidance that on my IHOF team, most if not all injuries happened to players with endurance below 50. Not counting the QB injuries as their endurance is hidden.

Yeah, that would seem to be the case. Next I'm going to do a study on that to see if that is what's happening. I was planning on doing it with this test, but I need a much larger sample size (than 5 per) to really get good results.