Pitcher inconsistency

[Spieler]

OOTP is based on season totals, has always been. Have a look, whatever league you are in (except the starter leagues). You have played 3742 games (until the next sims run), total runs are 16000 something, total hits are 33000 something, total HRs are 3600 something.

On top of that, your best and worst teams will be within a general range of each other. Best hitting team (I didn't chuck out the best and worst as outliers, but generally fits) scores 70% more runs than the worst, worst hitting team has 2x the Ks of the best team, etc. etc. Best team has ca. 130% more H+BB. In the leagues I looked at, the best were stacked with => gold players, the worst with <= bronze players.

Somehow, these numbers have to be spread out over a season. While the OOTP engine simulates a <balanced> league brilliantly, always did, there seems to be an issue where the player skills are unbalanced, either where all players in the league are really good, but also where teams are lopsided within a league.

Back in the Pursue-the-Pennant or Strat-O-Matic card playing days, if you stacked a team of "diamond" vs a team of "bronze", you would get stats that were realistic, but not within season parameters set by OOTP or MLB history. "Diamond" vs "diamond" would be (more) balanced, and more like "real" MLB historical totals.

The question then becomes, how are the stats balanced to meet league totals, especially in unbalanced leagues. You could build some theories, but obviously some randomness is involved. Hot&Cold can't be a factor, as is DH or not (not worth the coding). I have an idea, but can't back it up with stats. I did do a lot of tests back in the day (OOTP <=3) but unfortunately I don't have the data anymore.

But randomness is definitely a factor.

[Orcin]

Originally Posted by Spieler

OOTP is based on season totals, has always been. Have a look, whatever league you are in (except the starter leagues). You have played 3742 games (until the next sims run), total runs are 16000 something, total hits are 33000 something, total HRs are 3600 something.

On top of that, your best and worst teams will be within a general range of each other. Best hitting team (I didn't chuck out the best and worst as outliers, but generally fits) scores 70% more runs than the worst, worst hitting team has 2x the Ks of the best team, etc. etc. Best team has ca. 130% more H+BB. In the leagues I looked at, the best were stacked with => gold players, the worst with <= bronze players.

Somehow, these numbers have to be spread out over a season. While the OOTP engine simulates a <balanced> league brilliantly, always did, there seems to be an issue where the player skills are unbalanced, either where all players in the league are really good, but also where teams are lopsided within a league.

Back in the Pursue-the-Pennant or Strat-O-Matic card playing days, if you stacked a team of "diamond" vs a team of "bronze", you would get stats that were realistic, but not within season parameters set by OOTP or MLB history. "Diamond" vs "diamond" would be (more) balanced, and more like "real" MLB historical totals.

The question then becomes, how are the stats balanced to meet league totals, especially in unbalanced leagues. You could build some theories, but obviously some randomness is involved. Hot&Cold can't be a factor, as is DH or not (not worth the coding). I have an idea, but can't back it up with stats. I did do a lot of tests back in the day (OOTP <=3) but unfortunately I don't have the data anymore.

But randomness is definitely a factor.

Now we are getting somewhere. Thank you for this explanation.

[KleineBiere]

9 of 10 deGroms' will have a single season ERA between 3.00 & 4.96 (perfect level & min 162 IP)
9 of 10 Sale's will have a single season ERA between 3.03 & 5.53 (perfect level & min 162 IP)

4 of 5 Trout's (with 2400-3000 PA at PL) should have an OPS between .770 & .848

[Orcin]

Originally Posted by KleineBiere

4 of 5 Trout's (with 2400-3000 PA at PL) should have an OPS between .770 & .848

Our league has 10 of 11 within this range and the 11th is barely outside. Thanks for this confirmation that the variance is within the expected norm.

I think that, considering the two answers above, most of the complaints about poor results from certain cards are due to being on the low end of the expected range for a particular card or a sample size that is too small.

I still don't believe that the secret sauce has much to do with it. Great players (cards) make great managers (human players).

[DonkeyKongSr]

Originally Posted by Orcin

I still don't believe that the secret sauce has much to do with it. Great players (cards) make great managers (human players).

You seem to only care about information that supports your theory. What they are saying is absolutely true, but nothing said in those two posts explains how any player can remain at the high or low end of expectations for multiple seasons in a row. You are trying your hardest to justify that the secret to this game is almost entirely:

1) Get good players
2) Pray to the dice-rolling gods

To simplify it down to "good players make good managers" is insulting to anyone who has spent any significant time on this game beyond acquiring players.

[Leo_The_Lip]

Originally Posted by DonkeyKongSr

I don't claim to have the secret sauce, but isn't something along those lines a more logical explanation than luck why my Mike Trout has been better than every other Mike Trout for 6.5 straight leagues? Unless there is actually some hidden boost the developers are keeping a secret, it's virtually impossible for one card to be that consistently good on luck alone.

Anyway, he's on my second team which I've barely changed in those 6.5 seasons, so below is been pretty much the exact setup the entire time...

My experience is that having my best 'bunt for hit' players like Billy Hamilton set to extreme bunt for hit will, against extreme shift managers such as you, is like placing Billy Hamilton et al. on second base a lot.

So I don't think that is working at all.

[OMGPuppies]

Originally Posted by DonkeyKongSr

You seem to only care about information that supports your theory. What they are saying is absolutely true, but nothing said in those two posts explains how any player can remain at the high or low end of expectations for multiple seasons in a row. You are trying your hardest to justify that the secret to this game is almost entirely:

1) Get good players
2) Pray to the dice-rolling gods

To simplify it down to "good players make good managers" is insulting to anyone who has spent any significant time on this game beyond acquiring players.

Yup. I've witnessed countless times where pitchers with weak stats and ratings have career games and dominate my much better hitters. It's pretty frustrating.

[DonkeyKongSr]

Originally Posted by Leo_The_Lip

My experience is that having my best 'bunt for hit' players like Billy Hamilton set to extreme bunt for hit will, against extreme shift managers such as you, is like placing Billy Hamilton et al. on second base a lot.

So I don't think that is working at all.

I should probably decrease it a bit, but there are very, very few Billy Hamiltons or good Bunt for Hit players in the Diamond & Perfect leagues. There are Pull hitters galore.

EDIT: But, you know what, thanks a ton for reminding me to check this setting before each playoff series. I've lost to one or two teams in the playoffs that were more small ball focused and this setting may have hurt me quite a bit there.

[Bunktown Ballers]

Rating dont matter. I dont care if it's a 100 card, but a 76 performs better...go 76. Todd Helton Diamond card is garbage in Diamond league. Tony Oliva & Pete Rose in the 70's are bosses so far. Highest I've got is gold league with them .both are on pace for over 200 hits...Rod Carew 70+ something card too....rake like pro gardeners

[KleineBiere]

I was replaying the 2017 season 20 times


Starters with an ERA about 4.00 (deGrom at perfect)





A starter with a 4.00 career ERA & 0.6 SD will have an ERA between 2.82 & 5.18 in 19 of 20 seasons

[Maddox]

Originally Posted by zrog2000

This really isn't that hard to understand. These unbelievably great players are merely average in the advanced leagues, which leads to a huge variance in outcomes, just like for average players in a regular baseball season. I'm not sure where this idea that players should perform like they did in baseball history when they're all playing against each other came from. It simply is not possible for that to happen, because the stats just do not work out. If all the Babe Ruths and Barry Bonds hit 60-70 home runs and got on base 50% of the time, every deGrom wouldn't have a 1.something ERA because they're facing each other. But it makes sense that there would be a huge disparity in outcomes if deGrom is facing Babe Ruth 100 times and that sometime there would be 10 strikeouts in a row and other times there would be 5 home runs in 10 at bats. These are matchups between baseball gods.

Plus, it really seems like people have a hard time with what determines a meaningful sample size.

I think a season with over 600 plate appearances is a decent sample.

Here is my biggest problem with the argument that they are facing better pitcher/batter so you can't expect them to perform as well. It is that plenty of lesser hitters are facing them and doing better.

Look at the below graphic the 100 rated Mike Trout had the lowest BA on the team (with the exception of Kluszewski who only played a month).

Ozzie Smith, Keith Hernandez, Kevin McReynolds, Garry Maddox, Buddy Bell, Bob Boone and Luis Aparicio faced the same exact pitching. These are all lesser hitters than Trout yet they hit for better average. So when every single batter on my team hit for a higher average against the same Gold League pitchers (who are not all elite) I kind of expect more out of Trout.

If he had hit 250 I could shrug it off but .224 is hard to swallow. When Luis Aparicio's average is 34 points higher it is hard to swallow.

[DonkeyKongSr]

Originally Posted by Maddox

If he had hit 250 I could shrug it off but .224 is hard to swallow. When Luis Aparicio's average is 34 points higher it is hard to swallow.

What are the Avoid Ks on those other guys? Trout's only weakness is his strikeout rate, which leads to some lower AVG (less balls in play) at the higher levels. But, his OBP is still excellent thanks to his high-end walkrate.

Wait, the guy leading your team in HR and tied for the RBI lead is the guy you are talking about underperforming?

[Izz]

Originally Posted by Maddox

I think a season with over 600 plate appearances is a decent sample.

Here is my biggest problem with the argument that they are facing better pitcher/batter so you can't expect them to perform as well. It is that plenty of lesser hitters are facing them and doing better.

Look at the below graphic the 100 rated Mike Trout had the lowest BA on the team (with the exception of Kluszewski who only played a month).

Ozzie Smith, Keith Hernandez, Kevin McReynolds, Garry Maddox, Buddy Bell, Bob Boone and Luis Aparicio faced the same exact pitching. These are all lesser hitters than Trout yet they hit for better average. So when every single batter on my team hit for a higher average against the same Gold League pitchers (who are not all elite) I kind of expect more out of Trout.

If he had hit 250 I could shrug it off but .224 is hard to swallow. When Luis Aparicio's average is 34 points higher it is hard to swallow.

Wouldn't OBP be a better measure in this comparison, though? What's Aparicio's OBP compared to Trout's? My guess is Trout's OBP is around .350, whereas Aparicio's is below .300.

[uschi_baerchen]

Originally Posted by Orcin

Please explain for the ignorant people (referring to myself) among your audience.

It starts with "My Corbin (ERA to date 5,8) lost against his Corbin (ERA to date 2,9) 4-2. <but never vice-versa>, how come, that the same Card produces 5,8 and 2,9 sometimes" and it ends with: "It seems, that Corbin has an ERA around 4,0 but mine produced 4,8 last season and now he is at 5,5, I think I shall whack him".


(The following is life - data): 70 seasons of Corbin, average ERA 3,8, lowest ERA 2,4, 8 Season ERA less than 3, highest ERA 5,7, 6 Season ERA more than 5)

[uschi_baerchen]

Originally Posted by Orcin

That’s what I said.... dumb luck. You can’t control any of these factors, so whether your card hits a string of favorable or unfavorable is pure luck.

You are obiously incorrect. I offer you Sale or Corbin as a Pitcher with the added proviso, that you have to use the guy you choose for the next 20 years in your line-jup as a starter, I also tell you, that Corbins ERA last year was 2,25 and Sales ERA last year was 4,89. Which pitcher do you choose? Worse: why do you choose Sale?

[uschi_baerchen]

[QUOTE=Orcin;4439066]I am not a statistician, so I am asking those that are for an honest appraisal. Is this within the expected range of variance?



The math behind this is explained in



https://www.ootpdevelopments.com/boa...d.php?t=297089



You were actually involved in this discussion, I wonder, why you forgot?

[Zorro]

[QUOTE=uschi_baerchen;4440630]

Originally Posted by Orcin

I am not a statistician, so I am asking those that are for an honest appraisal. Is this within the expected range of variance?



The math behind this is explained in



https://www.ootpdevelopments.com/boa...d.php?t=297089



You were actually involved in this discussion, I wonder, why you forgot?

I don’t know about him but I can’t remember what I had for dinner last night let alone what post I read back in December.😁

[Markus Heinsohn]

Originally Posted by Orcin

I am not a statistician, so I am asking those that are for an honest appraisal. Is this within the expected range of variance?

I took the Trouts in my league that had over 2000 plate appearances (perfect leagues only). Surely that is a big enough sample size and yes there is variation in the competition etc.

To me, a variance in OPS between .841 and .702 for a card that is rated "perfect" is pretty large. However, that .702 could be an outlier. So if you tell me that the difference between the second-best and the second-worst OPS in this list is well within the expected results, then I will accept that this level of variance is normal and shut up.


Attachment 605457

This is perfectly fine.

Guys, this is all within normal statistical variances. There is no bug, it's the way it is meant to be and it is 100% realistic. It's all about sample size and the competition that the player faces.

We could easily eliminate all variances (if AVG < Expected_AVG then player_gets_hit but that would destroy the whole engine and be totally unrealistic.

I'll close this thread now, as it discusses a problem which does not exist.