Home | Webstore
Latest News: OOTP 25 Available - FHM 10 Available - OOTP Go! Available

Out of the Park Baseball 25 Buy Now!

  

Go Back   OOTP Developments Forums > Prior Versions of Our Games > Out of the Park Baseball 19 > OOTP 19 - Historical Simulations

OOTP 19 - Historical Simulations Discuss historical simulations and their results in this forum.

Reply
 
Thread Tools
Old 04-22-2018, 09:36 PM   #1
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
Pitcher usage in the pre-reliever era

This will follow-up on some of the observations I made in this thread. As a general matter, it is clear to me that OOTP has a difficult time accurately recreating pitcher usage in the "pre-reliever" era (roughly speaking, before 1940). In large part, this is due to the fact that the game's model is based on modern pitcher usage, which is very different from the way that pitchers were used prior to the development of relief specialists.

In order to get a better approximation of pitcher usage during this era, and to work within the limitations of the current game engine, I have suggested the following settings in the League Settings/Stats & AI page:
  • Pitching rotations should equal the number of pitchers on a team's staff or to "six," whichever is lower;
  • Rotation mode should be set to "start highest rested;"
  • Allow starters in relief;
  • Use of relievers and use of closers should be set to "very rarely."
Although this set-up is not historical (there were no set pitching rotations in that era), it seems to produce somewhat more realistic results.

Previously, I ran a replay of the 1922 season, the results of which are set out in above-linked thread. I wanted to see how this would work now that OOTP19 had introduced pitcher "hook" settings, so I did a replay of the 1916 season. In addition to the settings listed above, I set the number of relievers in the league to "2," thus giving each team an eight-man pitching staff on a 21-man roster. Eight pitchers may be bit high for 1916 (seven would probably be more realistic), but I wanted to see how this would work. The game's settings, in contrast, are:
  • 4-man pitching rotation;
  • Starting rotation mode is set to "default," which turns out to be "strict order;"
  • Number of relievers is set to "3;"
  • 25-man rosters.
I find these settings unsatisfactory for a number of reasons. No team in 1916 had a four-man rotation, which, of course, means that no team adhered to a strict rotation either. To begin with, as mentioned above, there were no pitching rotations in that period. More particularly, no team relied solely on four pitchers to carry the burden of starting games. Teams, in general, employed anywhere from eight to eleven starters over the course of the 1916 season, although Cleveland used thirteen. In addition, all of the pitchers on a staff were expected to start, and they picked up relief appearances on a more-or-less ad hoc basis (some pitchers were relief specialists, like Bernie Boland of the Tigers, but they were unusual). As a result, giving each team a three-man bullpen is completely unrealistic. Teams in that era typically didn't even have one-man bullpens, so three relievers per team is both excessive and a-historical. Furthermore, the AI usually designates one of these relievers to be the team's "stopper," which is also not how relievers were typically employed in that time period.

CONTINUED
joefromchicago is offline   Reply With Quote
Old 04-22-2018, 09:38 PM   #2
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
In my sim of the 1916 season, I chose to run the White Sox, although I made very minimal roster moves and only occasionally tinkered with the pitching rotation. I considered this to be a good test case, as the White Sox primarily used the same eight pitchers all season (Ed Walsh, the only other pitcher used by the Sox, appeared in just two games, once as a starter). Here are the numbers for the staff, compared with their real-life stats:

Reb Russell
(sim) 38 g, 31 gs, 266 ip
(RL) 56 g, 25 gs, 264.1 ip

Mellie Wolfgang
(sim) 37 g, 1 gs, 51.1 ip
(RL) 27 g, 14 gs, 127 ip

Lefty Williams
(sim) 35 g, 32 gs, 269 ip
(RL) 43 g, 26 gs, 224.1 ip

Joe Benz
(sim) 32 g, 30 gs, 249.1 ip
(RL) 28 g, 16 gs, 142 ip

Dave Danforth
(sim) 32 g, 0 gs, 48.1 ip
(RL) 28 g, 8 gs, 93.2 ip

Red Faber
(sim) 31 g, 30 gs, 252.2 ip
(RL) 35 g, 25 gs, 205.1 ip

Eddie Cicotte
(sim) 23 g, 22 gs, 171.2 ip
(RL) 44 g, 19 gs, 187 ip

Jim Scott
(sim) 8 g, 8 gs, 65 ip
(RL) 32 g, 21 gs, 165.1 ip

I moved Scott down to the number six spot in the rotation because he just wasn't pitching very well. That clearly skewed the numbers in favor of the other starters, particularly Joe Benz, who picked up 14 more starts in the sim than in real life (Benz ended up being my number-four starter). Otherwise, my changes to the rotation were minimal. In the sim, Williams pitched the most innings (269), while in real life Russell had the most (264.1), so I consider that fairly accurate.

The way that they accumulated their innings, however, was not. In the sim, the staff compiled 90 complete games, whereas in real life they had 73. Ironically, starters in the sim actually had more complete games, even though the game gave every team two pitchers who were designated as relievers. That is probably the result of a combination of the pitcher stamina ratings and strategy settings. I am not sure how the "hook" settings played a part in this, although presumably those played a role as well.

As might be expected with two men in the bullpen, the starters were used infrequently in relief. Reb Russell was the most versatile starter, appearing seven times in relief. That contrasts with 31 relief appearances for him in real life. In fact, Russel was used so often in relief that Baseball Reference classifies him as a relief pitcher, even though he had the second-most starts on the staff after Lefty Williams. That was, in fact, fairly typical of the era, and it's something that I saw in my 1922 replay, where the top starter was also used as the club's "high-leverage" reliever. Again, I'm not sure if that's because of the hook ratings or because of the additional man in the bullpen, but the fact that starters are being used less in relief represents something of a backwards step for the game in my opinion.

Even though I set rotation usage to "start highest rested," my experience with my 1922 replay was that the back end of the rotation would pick up a fair amount of starts because the top-end starters would be used frequently in relief. That didn't happen in my 1916 replay. Here's a comparison of the number of starts per pitcher in the sim and in real life:
  • (sim) 32-31-30-30-22-8-1-0
  • (RL) 26-25-25-21-19-16-14-8-1
So, at least in the case of the White Sox, starters started too often and relieved too infrequently. I'm not sure if setting rotation usage to "strict rotation, occasionally highest rested" would be an improvement, although I plan to test that. I'll also note that that result wasn't always the same with the AI teams. For instance, Pete Alexander led the majors with 45 starts in 1916, while in the sim he started 41 times. Likewise, Harry Coveleski of the Tigers started 39 games in real life and only 31 in the sim. In addition, for both the Philadelphia (N) and Detroit staffs, the top starters were used more often in relief (Alexander 10 times and Coveleski 7 times). That's something that warrants further testing.

CONTINUED

Last edited by joefromchicago; 04-22-2018 at 11:29 PM.
joefromchicago is offline   Reply With Quote
Old 04-22-2018, 09:42 PM   #3
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
One of the factors at play here may simply be the injury setting, which is set at "low." That's probably too low. There is a perception that players in the deadball era didn't get injured very often, which was not the case. It is probably true that players would more often play injured, but they'd get injured just as often as they do today. In 1916, for instance, Red Faber didn't pitch at all from June 4 to July 28 because he was sidelined with what the papers described as "rheumatism." The club didn't bring up anyone from the minors or make a trade. Instead, the rest of the staff was expected to carry the load (that's when Walsh picked up his only start of the season). So that meant some of Faber's starts were spread around to guys at the bottom of the staff like Danforth (three of his eight starts were in that period) and Wolfgang (seven of his fourteen starts).

That didn't happen in my sim. Faber was the only pitcher who suffered any kind of injury, and that resulted in a single trip to the fifteen-day disabled list and Wolfgang's sole start of the year. I suspect that AI teams with a more realistic distribution of starts saw injuries to one or more of their top-end starters, but my perception from reading my messages is that very few players ever spent any time on the disabled list.

Another common reason why pitchers in the bottom half of the staff picked up starts was because of frequent postponements which led to an astonishing amount of double-headers. From July 10 to July 20, the White Sox had a string of seven straight double-headers, all but one of them the result of rescheduled rain-outs. Even though those double-headers weren't all on consecutive days, playing fourteen games in eleven days certainly stretched out the pitching staff (which, remember, was missing Red Faber). In that span, the starting assignments were distributed like so: Williams-3, Russell-3, Scott-2, Benz-2, Cicotte-1, Wolfgang-2, Danforth-1. So the four pitchers with the fewest starts overall on the staff (Benz, Wolfgang, Cicotte, and Danforth) had six of the fourteen starts during this time, which is probably the kind of distribution we'd expect where a team was playing a bunch of games in a short period without one of its top starters.

It appears, however, that OOTP has something of the same problem with rainouts as it has with pitcher usage. In both cases, they're evidently set to modern standards. Today, because of CBA-mandated off-days and an aversion to double-headers, as well as state-of-the-art drainage systems, games in outdoor stadiums are rarely postponed because of weather. That wasn't the case a century ago, where double-headers were common and ballparks might be a swampy morass even a day after a heavy downpour. In 1916, the White Sox had 17 games rescheduled because of weather, 15 of which were rescheduled as the back ends of double-headers. In my sim, I reckon that I had two rainouts that resulted in double-headers.

So one under-appreciated factor in pitcher usage during the pre-reliever era is the frequency of double-headers, which are, in turn, largely the result of weather-related postponements. Another, albeit less-common factor is tied games. This is an extremely rare occurrence today in major league baseball, but it was far more prevalent in the era before lights were installed in ballparks. Today, a tied game that is called because of weather or a curfew is simply continued on another date. That's how OOTP handles that situation, but that was pretty much never done before the 1940s. Instead, a regulation game that was called with the score tied was simply entered into the books as a tied game. It didn't count in the standings, but the individual stats in the game did count. That's how the White Sox pitching staff, in 1916, compiled 155 total starts in a 154-game season (there was a 14-inning tied game against Washington on May 25 that was replayed on July 20, which was at the end of the team's streak of seven consecutive double-headers). Thus, a tied game will not only add a start to a pitcher's totals, but it will likely also add a double-header to the schedule, which will make it more likely that one of the back-end pitchers will get a start.

CONTINUED

Last edited by joefromchicago; 04-22-2018 at 09:51 PM.
joefromchicago is offline   Reply With Quote
Old 04-22-2018, 09:44 PM   #4
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
The biggest problem with the way OOTP handles pitchers in the pre-reliever era, however, is not with the way it handles starters but with the way it handles relievers. It was a big step forward when OOTP allowed starters to relieve. That, at least, solved one half of the problem. The other half, though, is that relievers aren't allowed to start (or, at least, they aren't allowed to be slotted in the rotation - they can still start on an emergency basis). The user can designate a reliever's primary or secondary role to be an emergency starter, but it doesn't appear that the AI ever does that. It would be an improvement if the AI would make those designations, but that doesn't really capture how most of these pitchers were actually used.

You could say that the White Sox had an emergency when they were looking at playing 14 games in 11 days without one of their top starters. But back-end starters were just as likely to pick up starts in non-emergency situations. Dave Danforth, for instance, had eight starts for the White Sox in 1916, but only two of them came in double-headers. He had starts in the first four months of the season, but after July 11 he was permanently relegated to the bullpen and it appears he was used mostly in a mop-up role thereafter. So he started out looking like a fifth or sixth starter, but for reasons unknown he dropped out and Joe Benz more-or-less took his place.

So was Danforth an emergency starter? Well, yes and no. In my sim, however, he was exclusively a reliever, appearing in 32 games without a start. That just didn't happen back in those days. Of the pitchers who appeared in 15 or more games in the majors that year, all of them started at least once. Bernie Boland of the Tigers led the majors with 37 relief appearances, but even he started nine times for Detroit. And though he finished 12 games that year, he trailed two other teammates - Jean Dubuc and George Cunningham - in that category. So it's not even accurate to say that Boland was the team's stopper, let alone its closer. Rather, it appears he was used in all sorts of situations - when the game was close, when it was out of reach, in the middle innings, in the late innings - it didn't seem to matter. His usage that year probably had more to do with the fact that Detroit had a pretty weak pitching staff and had to use everybody, even a guy like Boland who had an ERA+ of 73 (in my sim, Boland appeared 41 times, all in relief, and was assigned "stopper" as his primary role).

Was there anybody in 1916 who could be considered a reliever? Joe Oeschger of the Phillies had the most appearances (14) without a start in MLB that year. A look at his career, however, shows that he was definitely a starter, and his stats for 1916 can be explained by the fact that he was injured for much of the season. Slim Love of the Yankees, who had the highest percentage of relief appearances among all pitchers who appeared in at least 15 games (19 out of 20), was basically the last guy on the end of the bench, used increasingly in blowouts due to his inflated ERA. Two years later he started 29 games for the Yankees and led the AL in walks, and by 1921 he was out of majors. It's fair to say, then, that nobody was a true reliever in 1916, at least according to the current understanding of that term.

Contrast that with the results of my sim, which saw 13 pitchers with 20 or more appearances and with zero starts. Special mention should be made of Gene Packard, the MLB leader in appearances who trotted out to the mound 67 times for the Cubs with nary a start. Packard, in real life, started 16 times and relieved 21 times, with his starts scattered throughout the season. Nevertheless, Baseball Reference lists him as Chicago's closer (!) for 1916, which is simply ridiculous. Even OOTP has him as a middle reliever, although he went 11-11 with 11 saves, so I don't know what was going on with the Cubs' pitching staff. What I do know is that, historically, no one recorded 67 or more relief appearances in a season until 1943, when Ace Adams did it for the Giants. Adams could genuinely be considered a stopper. Packard was just a fifth starter who relieved when needed.

Why, then, does OOTP designate pitchers as relievers in the pre-reliever era? Again, it is largely due to the way that OOTP models the game on how it is currently played. Today, there is a pretty sharp distinction between relievers and starters. Of the 122 pitchers in the majors in 2017 who had 20 or more starts, only six also had ten or more relief appearances. Likewise, of the 272 pitchers who had 20 or more relief appearances, only five had ten or more starts. In 1916, in contrast, of the 60 pitchers who started 20 or more games, 36 relieved in at least ten, and of the 21 pitchers who had 20 or more relief appearances, all but five of them started at least ten times. It's pretty clear, then, that pitchers in the pre-reliever era were generally expected to start and relieve. That does not mesh with OOTP's model, which is based on the way that pitchers are used today. Imposing that model on the pre-reliever era, however, results in an awkward fit.

CONTINUED
joefromchicago is offline   Reply With Quote
Old 04-22-2018, 09:47 PM   #5
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
What, then, is to be done?

Ideally, the entire OOTP pitching model should be scrapped and something more flexible should be put in its place. I calculate the odds of that happening at around 0%. Realistically, then, any solution would have to fit into the current pitching model. With that in mind, here are some suggestions:

(1)(a) Expand pitching rotation limits: currently, pitching rotations cannot exceed six. In practice, that means that any pitcher in excess of the rotation limit is automatically considered a reliever, even if, in all other respects, he should be a starter. Consequently, in order to give starts to the bottom half of the pitching staff in the pre-reliever era, the pitching rotation should be equal to the number of pitchers on the staff, so that every pitcher is in the rotation (and then every pitcher would also be allowed to relieve). Or, in the alternative:

(1)(b) Allow relievers to start: If pitching rotations cannot be expanded to encompass the entire pitching staff, then relievers should be allowed to start in the same way that starters are now allowed to reliever. And that means more than just designating some relievers as emergency starters (which apparently the AI doesn't do anyway). Under the current set-up, most relievers should have a primary role as "starter," and only a secondary role as a reliever. There would only be a few select cases where a relief pitcher would be exclusively a reliever (like Fred "Firpo" Marberry in 1925) or even primarily a reliever (like Bernie Boland in 1916). This would help to blur the sharp distinction between starters and relievers which is not appropriate for the pre-reliever era.

(2) Introduce the role of "relief specialist:" The best approximation of reliever usage in the pre-reliever era, according to OOTP terminology, would be "middle reliever." That is to say, a reliever who is brought into the game in the middle innings in a non-save situation. To be sure, that's not the only type of situation in which relievers were used, but it's a far better approximation than either "closer" or "stopper." Better yet, OOTP should introduce a new reliever role, that of "relief specialist." A relief specialist would be a pitcher who, although he can start, is used primarily in an all-purpose relief role. That would account for pitchers who might pick up a start here or there but who would be used mainly in "low-leverage" relief situations.

(3) Set closer usage to "never:" In conjunction with (2) above, eliminate all closers and stoppers in the pre-reliever era, except for exceptional cases (Marberry, again). Right now, the lowest level that the gamer can set for closer usage is "very rarely," which is still too often, and which the AI evidently doesn't follow anyways.

(4) Set some starters' secondary role to "high leverage:" In my 1922 replay, it seemed like the AI was doing this automatically. So the number-one or number-two starter in the rotation would double as the team's high-leverage reliever. I didn't see that, however, in my 1916 replay, and the AI apparently never assigns a secondary role to any member of the starting rotation. In history, there were plenty of instances where a top-half starter would also rack up a lot of relief appearances. It wasn't necessarily the guy at the top of the staff, but would often be the guy with the second- or third-most starts. Pitchers like Three-Finger Brown and Ed Walsh and even Lefty Grove are known primarily as starters but who made quite a few relief appearances as well. That sort of dual role should be possible in the game.

(5) Introduce era-specific standards for postponements: As explained above, weather-related postponements appear to be calibrated to current conditions, which means that there are too few of them in the pre-modern era. This function, however, could probably be handled with a mod, so changing the game might not be necessary.

(6) Allow for tied games: This has been suggested in the past, usually in connection with the Japanese and Korean leagues, which permit tied games. It would also be useful for leagues in the pre-reliever era, as tied games were not uncommon prior to the introduction of lights.
joefromchicago is offline   Reply With Quote
Old 04-22-2018, 10:02 PM   #6
mitchkenn
Hall Of Famer
 
mitchkenn's Avatar
 
Join Date: Jan 2015
Location: Oregon, not by design
Posts: 2,853
WOW! what a write up. a lot of that stuff i sorta knew, but not really as much as i thought after reading your missive. And i honestly liked your suggestions at the end on how to modify the game for era specific roles. Good stuff and hopefully the powers-that-be will use them in upcoming versions or at least consider them very sincerely.
__________________
"This is my opening farewell " - Jackson Browne
“They make a desolation and call it peace.” ― Agha Shahid Ali
"Maybe she just has to sing, for the sake of the song - And who do I think that I am to decide that she's wrong." - Townes Van Zandt
"I saw a young man leaning on his wooden crutch - He called out to me, 'Don't ask for so much' And a young woman leaning in her darkened door She cried out to me, 'Why not ask for more?' " - Leonard Cohen
"Hello darkness, my old Friend ...." - Paul Simon
Before Mays, before DiMaggio, there was Oscar Charleston.
"All the lies about Babe Ruth are true." - Waite Hoyt

Avatar is the late great Townes Van Zandt. rip.
mitchkenn is offline   Reply With Quote
Old 04-22-2018, 10:10 PM   #7
Orcin
Hall Of Famer
 
Orcin's Avatar
 
Join Date: Aug 2010
Location: Indiana
Posts: 9,798
Great series of posts. Having just started to play the deadball era this year, I have a much better appreciation for these ideas and why they are needed.
Orcin is offline   Reply With Quote
Old 04-23-2018, 03:58 PM   #8
JaBurns
All Star Starter
 
Join Date: Sep 2015
Location: Kelowna, British Columbia
Posts: 1,258
Thank you for the insights into the dead ball era pitching. I had heard some a those facts mentioned piecemeal but it is nice to have them all in one place. Your suggestions are sound, except I don’t like the idea of allowing tie games. But that’s just me take it with a grain of salt.
JaBurns is online now   Reply With Quote
Old 04-24-2018, 08:53 AM   #9
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
Thanks everyone for your comments.

Quote:
Originally Posted by JaBurns View Post
Your suggestions are sound, except I don’t like the idea of allowing tie games.
Maybe because you've never seen one Tied games in MLB are about as common as hens' teeth, although there was one at the tail end of the 2016 season between the Cubs and Pirates. As Retrosheet describes it:
Quote:
CUBS 6TH: Almora grounded out (shortstop to first); Game called for rain; the game was not suspended and resumed later because of its proximity to the end of the season and that it had no bearing on the standings; 0 R, 0 H, 0 E, 0 LOB. Cubs 1, Pirates 1.
The way that MLB and OOTP handle suspended games was only introduced in the 1940s. The concern back then was that a team would turn off the lights in its ballpark to stop a game that it was losing and force a do-over later. That procedure was later extended to all games, whether day or night, although I'm not entirely sure when that happened.

So OOTP's method of handling suspended games will guarantee that there are no tied games in the modern game. But it's anachronistic to apply those rules to leagues in the pre-lights era.

Last edited by joefromchicago; 04-24-2018 at 09:04 AM.
joefromchicago is offline   Reply With Quote
Old 04-24-2018, 09:28 AM   #10
italyprof
All Star Reserve
 
Join Date: Feb 2014
Posts: 977
JoefromChicago, brilliant historical research. Painstaking work on your part on all of our behalf, and a great set of suggestions. Many thanks.

I notice that you did not note the new function under Stats and AI that allow for slower or faster than "normal" pulling of starters and relievers. My guess is that for pre-relief pitcher eras we should put each to the maximum slow pull setting, no? Though I admit I don't know the actual effect yet.
italyprof is offline   Reply With Quote
Old 04-24-2018, 10:40 AM   #11
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
Quote:
Originally Posted by italyprof View Post
I notice that you did not note the new function under Stats and AI that allow for slower or faster than "normal" pulling of starters and relievers. My guess is that for pre-relief pitcher eras we should put each to the maximum slow pull setting, no? Though I admit I don't know the actual effect yet.
I honestly have no clue what the hook settings mean. The 1916 settings are -3 (between quick and very quick) for starters and +2 (slow) for relievers. But what does that mean? When starters were completing over 50% of their starts, it's clear managers weren't routinely pulling their starters quickly in 1916. And how do the hook settings interact with the "Use of Relievers" setting? Frankly, I'm baffled.

In my 1916 replay, starters were still completing most of their starts (indeed, they may have been completing too many), so it doesn't look like they were getting quick "hooks." In the replay, the White Sox pitching staff compiled 238 total appearances, compared with 298 appearances for the Sox in real life. That's a fairly significant difference. Ironically, it appears that the White Sox in 1916, which didn't have any dedicated relievers, used relievers more often than they did in my replay, in which the team had two pitchers who were designated exclusively as relievers. On the other hand, the Cubs, which had the ML leader in appearances in my replay, compiled 292 total appearances while, in real life, they had 275. So the use of relievers may have more to do with the quality of the starters than with the hook settings.

I'm not sure what to make of all this. I'll note that my settings were not the same as the game's default settings for 1916, so that might explain some of the statistical deviation. But the default setting would give each club a four-man rotation and three-man bullpen. My guess is that those staffs might yield more overall appearances, but the distribution of starts would certainly be more skewed.
joefromchicago is offline   Reply With Quote
Old 04-26-2018, 10:19 AM   #12
David Watts
Hall Of Famer
 
David Watts's Avatar
 
Join Date: Apr 2002
Location: Looking for a place called Leehofooks
Posts: 8,817
Very interesting thread. Fired up 1905 last night and set it up so teams used 6 starters, zero relievers, always start the highest rested. Early in the season, but I am liking what I'm seeing.

Wonder how hard it would be for Markus to up rotation size to 7 or even 8?
David Watts is online now   Reply With Quote
Old 04-26-2018, 11:05 AM   #13
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
Quote:
Originally Posted by David Watts View Post
Wonder how hard it would be for Markus to up rotation size to 7 or even 8?
That seems like the simplest and best solution to me. And I don't think it would be necessary to extend the limit beyond eight. By the time pitching staffs expanded to nine or more pitchers, teams were already employing relief specialists and they were gradually adopting true rotations.

I don't know how difficult it would be to expand the rotation limit. Unfortunately, one reason I think the developers might hesitate to do that would be their fear of the OOTP community's reaction to such a change. I can imagine people on these boards saying: "Eight-man rotations? You gotta' be kidding me! That's so unrealistic. Who has ever done that?" And the answer would be: "Nobody has ever done that. But as long as OOTP is stuck with this rotation-based model of pitching staffs, that's the best way to model pitcher usage in the pre-reliever era."
joefromchicago is offline   Reply With Quote
Old 04-26-2018, 11:41 AM   #14
David Watts
Hall Of Famer
 
David Watts's Avatar
 
Join Date: Apr 2002
Location: Looking for a place called Leehofooks
Posts: 8,817
Quote:
Originally Posted by joefromchicago View Post
That seems like the simplest and best solution to me. And I don't think it would be necessary to extend the limit beyond eight. By the time pitching staffs expanded to nine or more pitchers, teams were already employing relief specialists and they were gradually adopting true rotations.

I don't know how difficult it would be to expand the rotation limit. Unfortunately, one reason I think the developers might hesitate to do that would be their fear of the OOTP community's reaction to such a change. I can imagine people on these boards saying: "Eight-man rotations? You gotta' be kidding me! That's so unrealistic. Who has ever done that?" And the answer would be: "Nobody has ever done that. But as long as OOTP is stuck with this rotation-based model of pitching staffs, that's the best way to model pitcher usage in the pre-reliever era."
I really hope this thread catches the eye of Markus. You have some wonderful ideas here.
David Watts is online now   Reply With Quote
Old 04-26-2018, 04:55 PM   #15
JaBurns
All Star Starter
 
Join Date: Sep 2015
Location: Kelowna, British Columbia
Posts: 1,258
I think this thread has Markus’ attention he has already said changes are going to be made to the draft setup.
JaBurns is online now   Reply With Quote
Old 04-26-2018, 04:57 PM   #16
JaBurns
All Star Starter
 
Join Date: Sep 2015
Location: Kelowna, British Columbia
Posts: 1,258
Quote:
Originally Posted by joefromchicago View Post
That seems like the simplest and best solution to me. And I don't think it would be necessary to extend the limit beyond eight. By the time pitching staffs expanded to nine or more pitchers, teams were already employing relief specialists and they were gradually adopting true rotations.

I don't know how difficult it would be to expand the rotation limit. Unfortunately, one reason I think the developers might hesitate to do that would be their fear of the OOTP community's reaction to such a change. I can imagine people on these boards saying: "Eight-man rotations? You gotta' be kidding me! That's so unrealistic. Who has ever done that?" And the answer would be: "Nobody has ever done that. But as long as OOTP is stuck with this rotation-based model of pitching staffs, that's the best way to model pitcher usage in the pre-reliever era."
Then don’t call it a “rotation” just call it pitching staff.
JaBurns is online now   Reply With Quote
Old 04-27-2018, 01:49 AM   #17
Le Grande Orange
Hall Of Famer
 
Le Grande Orange's Avatar
 
Join Date: Feb 2002
Location: Up There
Posts: 15,386
Quote:
Originally Posted by joefromchicago View Post
So one under-appreciated factor in pitcher usage during the pre-reliever era is the frequency of double-headers, which are, in turn, largely the result of weather-related postponements.
Two thoughts on that:

(1) This is why, for any test of the game against historical stats such as you are doing, I would suggest using the as played historical schedules, since these include the off days and doubleheaders caused by postponements. If you get closer-to-history results using those, that is good evidence the schedule is having a measurable impact on results.

(2) Games were postponed in far greater numbers back then than is the case in the contemporary era, and were also rescheduled differently. (Clubs frequently did not complete their full schedule as some games were inevitably cancelled and not made up.) OOTP does not recreate the trend toward fewer postponements over time, it does not recreate the historical postponement rates, nor does it recreate the rescheduling practices of the past. Without such recreations, there is always going to be issues in comparisons.

Quote:
Originally Posted by joefromchicago View Post
Another, albeit less-common factor is tied games. This is an extremely rare occurrence today in major league baseball, but it was far more prevalent in the era before lights were installed in ballparks.
True, but there were still a tiny portion of all games played. In 1916, there were 20 tie games (out of a nominal 1232 in the schedule, or about 1.6%).

Quote:
Originally Posted by joefromchicago View Post
Today, a tied game that is called because of weather or a curfew is simply continued on another date. That's how OOTP handles that situation, but that was pretty much never done before the 1940s.
Suspended games didn't exist until the rules regarding them were first invented, which was the latter half of the 1940s, if I recall correctly. And even then, the suspended game rules evolved over time.
Le Grande Orange is offline   Reply With Quote
Old 04-27-2018, 09:21 AM   #18
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
Quote:
Originally Posted by Le Grande Orange View Post
(1) This is why, for any test of the game against historical stats such as you are doing, I would suggest using the as played historical schedules, since these include the off days and doubleheaders caused by postponements. If you get closer-to-history results using those, that is good evidence the schedule is having a measurable impact on results.
That's a fair point that I hadn't considered. I'll make that one of my future tests. My focus, however, is not just on historical replays but also on fictional leagues that aim to recreate conditions in the pre-reliever era. Those happen to be the kinds of leagues that I'd prefer to play, and they don't have as-played schedules.

Quote:
Originally Posted by Le Grande Orange View Post
Suspended games didn't exist until the rules regarding them were first invented, which was the latter half of the 1940s, if I recall correctly. And even then, the suspended game rules evolved over time.
Yeah, I'm pretty sure I said all that.

Last edited by joefromchicago; 04-27-2018 at 11:47 AM.
joefromchicago is offline   Reply With Quote
Old 04-27-2018, 04:28 PM   #19
Le Grande Orange
Hall Of Famer
 
Le Grande Orange's Avatar
 
Join Date: Feb 2002
Location: Up There
Posts: 15,386
Quote:
Originally Posted by joefromchicago View Post
Yeah, I'm pretty sure I said all that.
Well, there is a difference between true suspended games, that is, the specific rules for such, and games which became de facto suspended games as a result of a club successfully protesting a game, and the game ordered continued from the point of the protest.
Le Grande Orange is offline   Reply With Quote
Old 04-27-2018, 11:35 PM   #20
joefromchicago
Hall Of Famer
 
joefromchicago's Avatar
 
Join Date: Jun 2011
Posts: 3,630
Quote:
Originally Posted by Le Grande Orange View Post
Well, there is a difference between true suspended games, that is, the specific rules for such, and games which became de facto suspended games as a result of a club successfully protesting a game, and the game ordered continued from the point of the protest.
I would be very interested in reading your thoughts regarding the substance of my posts.
joefromchicago is offline   Reply With Quote
Reply

Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 04:25 PM.

 

Major League and Minor League Baseball trademarks and copyrights are used with permission of Major League Baseball. Visit MLB.com and MiLB.com.

Officially Licensed Product – MLB Players, Inc.

Out of the Park Baseball is a registered trademark of Out of the Park Developments GmbH & Co. KG

Google Play is a trademark of Google Inc.

Apple, iPhone, iPod touch and iPad are trademarks of Apple Inc., registered in the U.S. and other countries.

COPYRIGHT © 2023 OUT OF THE PARK DEVELOPMENTS. ALL RIGHTS RESERVED.

 

Powered by vBulletin® Version 3.8.10
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright © 2020 Out of the Park Developments