When people criticize OOTP for being unrealistic...

[kriscolic]

there's a knee-jerk reaction on these forums, I've noticed. Small sample size caveats and selective memory accusations abound. But maybe, just maybe, when the simulation spits out, say, a 45 run performance from a team, it's evidence that something is amiss. Literally a result that would never happen.

[Will_L]

In 2007 the Rangers tore the Orioles a new one by a score of 30-3.
Here's a link to a minor league shocker...
https://www.sportingnews.com/us/mlb/...11w4tdw7unj1cu

[ALB123]

Originally Posted by Will_L

Here's a link to a minor league shocker...
https://www.sportingnews.com/us/mlb/...11w4tdw7unj1cu

That game was probably still shorter than your average Red Sox v Yankees game. n'yuck, n'yuck, n'yuck

[RchW]

Apples and Oranges. In real life that doesn’t happen yet because the humans don’t let it. No one could program the number of interactions required to temper outlier scores realistically. Should OOTPB stop at a number, 23, 30?

Knees never felt better.��

[kriscolic]

Originally Posted by Will_L

In 2007 the Rangers tore the Orioles a new one by a score of 30-3.
Here's a link to a minor league shocker...
https://www.sportingnews.com/us/mlb/...11w4tdw7unj1cu

45 is not 30. It's literally half again as much. And baseball in 1869 was a wee bit different.

OOTP does a great job of getting things right in the aggregate, but the distribution can be a bit peculiar. Improbabilities are more probable than they should be. Like the time my team scored exactly 20 runs in three straight games. Or the time my shortstop made 4 throwing errors in the span of 6 batters. I could go on. We all could. The distribution of events is important to the realism of the simulation. A team going 81-81 while scoring 810 runs and allowing 810 runs makes sense. If they win half their games 10-0 and lose the other half 0-10, though, you've got a problem.

[CBeisbol]

Originally Posted by kriscolic

Improbabilities are more probable than they should be.

I don't know that this is true

Improbabilities have a lot more chance to happen in OOTP because there are thousands, or millions, more universes.

Of course unlikely events will happen more




Edit
https://www.nytimes.com/2009/06/30/s...ll/30bats.html

Originally Posted by kriscolic

. Or the time my shortstop made 4 throwing errors in the span of 6 batters. I

https://youtu.be/qMQJya1AyXQ

[joefromchicago]

Originally Posted by kriscolic

there's a knee-jerk reaction on these forums, I've noticed. Small sample size caveats and selective memory accusations abound. But maybe, just maybe, when the simulation spits out, say, a 45 run performance from a team, it's evidence that something is amiss. Literally a result that would never happen.

Translation: "Let me post a line score without any surrounding context that might explain the seemingly aberrant result and then complain that it's evidence that the game is broken."

Strange results like this can be caused by a number of factors. In my experience, this happens a lot when position players pitch, either because they're inserted into a game that's already a blowout or because there aren't enough pitchers on the staff to begin with (a common problem in lower minor leagues, such as the GCL). That, in turn, is the result of either the gamer not paying attention to rosters in the low minors or else relying on historical rosters even where those rosters are incomplete or inadequate. I'd rank that as 50% the game's fault and 50% the gamer's fault.

Here, we can get a pretty good idea what happened with the small bits of evidence found in the screen cap. The game recap says that the key hits were off of pitchers M. Thompson, R. Mercedes, and N. Shoaf. The Red Sox GCL franchise in 2005 didn't have anyone named Thompson, Mercedes, or Shoaf on its roster. Indeed, the Red Sox didn't have anyone in their entire minor league organization with those last names. A bit of detective work, however, can solve that puzzle.

No major leaguer has ever been named Shoaf, but a "Nick Shoaf" played five games as a relief pitcher for the Kansas City T-Bones of the independent Northern League in 2005. That must be the "N. Shoaf" who gave up a three-run homer in the top of the eighth. Firing up OOTP and creating a 2005 game with historical minors, it turns out that "R. Mercedes" is Roque Mercedes, a reliever in the Brewers' organization who never made it above AA. There were three Mike (Michael) Thompsons playing pro ball in 2005. My guess is that the M. Thompson who gave up five runs in the top of the first was Mike Thompson, who pitched briefly for the Braves' GCL club.

All three of those players were pitchers, it's true, but they were all bad pitchers and they were all relievers. The fact that Thompson, a two-pitch reliever with a low stamina, started the game is a red flag that indicates something isn't right. Either every starter was tired, or else there were no starters on the team - a common enough problem in the low minors, but one that can be addressed through some basic roster management on the gamer's part.

So what can we conclude here? A game in the lowest rung of the minors was a blow-out because one team didn't have good pitching. I'd venture to guess that a large majority of OOTP players wouldn't even care about that kind of result, as their focus is solely on developing players in the minors and getting them playing time, regardless of the final score. For those people who really care about getting accurate results for their Gulf Coast League games, they should be prepared to put more time and effort into getting the league settings right and monitoring the league for any imbalances among the rosters.

[CBeisbol]

Originally Posted by joefromchicago

So what can we conclude here? A game in the lowest rung of the minors was a blow-out because one team didn't have good pitching. I'd venture to guess that a large majority of OOTP players wouldn't even care about that kind of result, as their focus is solely on developing players in the minors and getting them playing time, regardless of the final score. For those people who really care about getting accurate results for their Gulf Coast League games, they should be prepared to put more time and effort into getting the league settings right and monitoring the league for any imbalances among the rosters.

Sounds like a problem with the game. The AI doesn't handle the low minors well

[joefromchicago]

Originally Posted by CBeisbol

Sounds like a problem with the game. The AI doesn't handle the low minors well

I wouldn't necessarily place all of the blame on OOTP. I think a lot depends on how much the gamer wants realistic game results from the low minors. I do think there's an imbalance between the number of starters and relievers in the game, where there are too many of the latter and not enough of the former. For historical minors, that's definitely a game problem. For fictional minors, the gamer can always fix the problem by simply creating more starters. There are options to fill real minor-league rosters with fictional or ghost players - I'm not sure how well that works for the lowest rung of the minors which often has no roster limits, but that's another option to look at.

Bottom line: if the gamer really doesn't want to see teams in the GCL score 45 runs in a game, there are steps that the gamer can take to minimize that possibility.

[Brad K]

Originally Posted by CBeisbol

Improbabilities have a lot more chance to happen in OOTP because there are thousands, or millions, more universes.

Additionally there are lots of OOTP players who have simulated more games than have been played in baseball history.

[itsmb8]

Originally Posted by joefromchicago

I wouldn't necessarily place all of the blame on OOTP. I think a lot depends on how much the gamer wants realistic game results from the low minors. I do think there's an imbalance between the number of starters and relievers in the game, where there are too many of the latter and not enough of the former. For historical minors, that's definitely a game problem. For fictional minors, the gamer can always fix the problem by simply creating more starters. There are options to fill real minor-league rosters with fictional or ghost players - I'm not sure how well that works for the lowest rung of the minors which often has no roster limits, but that's another option to look at.

Bottom line: if the gamer really doesn't want to see teams in the GCL score 45 runs in a game, there are steps that the gamer can take to minimize that possibility.

The entire problem with ALL baseball video games. There are very little, if not zero, pitchers that are strictly a reliever. They're either a starter that isnt good enough or cant eat up innings well that is moved to the bullpen, or theyre a closer that isnt good enough that becomes a set-up guy. No pitcher is ever a strict "early-mid inning reliever" type of pitcher.

I wish baseball video games would recognize this and list pitchers as only starters or closers, and develop the rotation/bullpen management around that.

[mytreds]

Happening one time is realistic.

Happening all the time would be a stretch I agree.

Outliers exist. What happened to you was exceptional. Foolish Baseball would probably make a good video out of it had it happened in real life XD

[Bunktown Ballers]

Originally Posted by RchW

Apples and Oranges. In real life that doesn’t happen yet because the humans don’t let it. No one could program the number of interactions required to temper outlier scores realistically. Should OOTPB stop at a number, 23, 30?

Knees never felt better.��

I think I reacted this same way 20 years ago. Has happened again since & chances are neither will you. I remember in real life a pitcher came into game, faced 1 batter, got ejected, new pitcher came in and pitched a perfect game thereafter (didn't count as Perfect game)....stuff happens

[joefromchicago]

Originally Posted by itsmb8

The entire problem with ALL baseball video games. There are very little, if not zero, pitchers that are strictly a reliever. They're either a starter that isnt good enough or cant eat up innings well that is moved to the bullpen, or theyre a closer that isnt good enough that becomes a set-up guy. No pitcher is ever a strict "early-mid inning reliever" type of pitcher.

I wish baseball video games would recognize this and list pitchers as only starters or closers, and develop the rotation/bullpen management around that.

I think that's basically correct. Nobody goes into pro ball thinking "I'm going to be the best middle reliever ever!" I'd go further and say that most closers probably began as starters as well. Mariano Rivera, for instance, had 68 starts in the minors and 10 starts in his first season with the Yankees before being converted into a full-time closer.

Teams draft pitchers, not relievers and starters, and there's a premium on starters - after all, they pitch over 50% of a team's innings but only represent about 40% of a team's staff. Just from a cost/benefit perspective, therefore, a team gets more "bang for its buck" when it develops a starter.

I'd agree, then, that relievers are, for the most part, guys who couldn't cut it as starters. OOTP doesn't do a very good job reflecting that fact. In my experience, the distinction between starters and relievers is set at the draft and doesn't change much in the minors. That's not realistic.

[ALB123]

Originally Posted by joefromchicago

I think that's basically correct. Nobody goes into pro ball thinking "I'm going to be the best middle reliever ever!" I'd go further and say that most closers probably began as starters as well. Mariano Rivera, for instance, had 68 starts in the minors and 10 starts in his first season with the Yankees before being converted into a full-time closer.

That raises a question I've had. If you take a Mariano Rivera or a Trevor Hoffman type of famous closer and have him start games in GCL or Short-Season A leagues, will that negatively impact their development in OOTP? My belief is to let them get innings on that arm in any way possible as long as you're not beating on them to pitch more and more, not wait for Save opportunities and an occasional non-Save inning or two in Rookie Level games.

Thoughts, anyone?

[Syd Thrift]

I don’t think the game gives debuffs for playing starters as relievers per se. The main factors in a player developing are:

- Playing time
- Not being too bad or too good for the level
- The player’s mental attributes
- The attributes of his coaches
- Your team’s development budget (although the effects of this are very small)

If you put a starter into a crowded bullpen they might not play as much and might hurt their development that way, or if for instance they throw a lot of mediocre pitches instead of 2 good ones they might not pitch as effectively as they would if they were starters, which could also ****** their development. But if they’re still playing a lot, I don’t think stamina for instance develops any differently.

[BIG17EASY]

Originally Posted by joefromchicago

I think that's basically correct. Nobody goes into pro ball thinking "I'm going to be the best middle reliever ever!" I'd go further and say that most closers probably began as starters as well. Mariano Rivera, for instance, had 68 starts in the minors and 10 starts in his first season with the Yankees before being converted into a full-time closer.

Actually, Rivera was John Wetteland's setup man in 1996 and worked a career-high 107.2 innings. But your point is right about him being a starter until he couldn't hack it as a starter.

The interesting thing about Rivera (and I'm going off on a tangent here) is that his K/9 rate went down when he became a closer, where as most guys see their K/9 rate go up when they move from starter to ace reliever because they are relying on one or two overpowering pitches in a short stint and they can air out the fastball for 15-20 pitches rather than pacing themselves for multiple innings. In '96 as a setup man, Rivera's K/9 was 10.9 (130 Ks in 107.2 innings). In the next 17 years as a closer, his highest K rate was 9.8 in 2008 and 2009, and he posted a K rate of 9.0 or better just five times in those 17 seasons. It's a testament to just how good he was that he had a higher contact rate than most closers, yet was still nearly untouchable. And it also speaks to some of his postseason failures due to not being able to get a swing-and-miss all the time. (Yes, I do realize a guy with 96 postseason appearances will have more postseason failures than anyone else, so don't come at me, Yankee fans.)

[itsmb8]

Originally Posted by joefromchicago

I think that's basically correct. Nobody goes into pro ball thinking "I'm going to be the best middle reliever ever!" I'd go further and say that most closers probably began as starters as well. Mariano Rivera, for instance, had 68 starts in the minors and 10 starts in his first season with the Yankees before being converted into a full-time closer.

Teams draft pitchers, not relievers and starters, and there's a premium on starters - after all, they pitch over 50% of a team's innings but only represent about 40% of a team's staff. Just from a cost/benefit perspective, therefore, a team gets more "bang for its buck" when it develops a starter.

I'd agree, then, that relievers are, for the most part, guys who couldn't cut it as starters. OOTP doesn't do a very good job reflecting that fact. In my experience, the distinction between starters and relievers is set at the draft and doesn't change much in the minors. That's not realistic.

What would happen if you changed every RP's position in the game to SP? Im curious as to what would happen with roster, rotation, and bullpen logic.

[ALB123]

Originally Posted by Syd Thrift

I don’t think the game gives debuffs for playing starters as relievers per se.

Would you say the same is true about playing Relievers as Starters when developing?

[Thundercrack]

Originally Posted by CBeisbol

Improbabilities have a lot more chance to happen in OOTP because there are thousands, or millions, more universes.





Edit
https://www.nytimes.com/2009/06/30/s...ll/30bats.html


https://youtu.be/qMQJya1AyXQ

somehow I never thought about that. makes sense there would be a wide variance