why are tied games in the historical record not replayed

[uruguru]

If I am playing with historical transaction and lineups, I have noticed that games that ended tied but official are not in the "actual played" schedule.

Lineups happened, pitchers made starts.... the stats count. Is there a reason why they are not replayed in the historical game?

[joefromchicago]

Because they were tied games. If every game (scheduled and make-up) was included, you'd end up playing too many games to a result.

[uruguru]

Originally Posted by joefromchicago

Because they were tied games. If every game (scheduled and make-up) was included, you'd end up playing too many games to a result.

These games were not made up, though. A team might have 160 games played on its schedule now. Is that not a problem? It feels like it might be.

[Rain King]

It would probably be helpful to provide an example.

[uruguru]

Originally Posted by Rain King

It would probably be helpful to provide an example.

This tie game was never made up.

https://www.baseball-reference.com/b...96209090.shtml

These two teams also had another rainout that was never rescheduled, so both teams ended up with 160 games for the season.

[AESP_pres]

Because they didn't count and had no influence on the standings... playing them with a winning team would be rewriting history.

[uruguru]

Originally Posted by AESP_pres

Because they didn't count and had no influence on the standings... playing them with a winning team would be rewriting history.

what do you mean, rewriting history? isn't that what you are doing every single time you run a historical sim?

It was just a simple question. It's a missing game on OOTP's "actual played" schedule that was actually played and never made up, so restoring it doesn't push anyone over a full schedule. There are lineups and starting pitchers.

[David Watts]

Pretty sure this isn't unique to OOTP and all replay games do it this way.

[uruguru]

Originally Posted by David Watts

Pretty sure this isn't unique to OOTP and all replay games do it this way.

well that makes it ok then, I guess. I mean, they literally have a person dedicated to getting everyone's hometown correct. Why not just have someone fix the handful of tie games that were actually played and never made up? less effort. I'll go find them all if they need someone to do it.

[AESP_pres]

Originally Posted by uruguru

what do you mean, rewriting history? isn't that what you are doing every single time you run a historical sim?

Yes and no... Yes we changed the result of the games (even playing with real transaction / real lineup like I do) but... they were played games, tied games don't count as played in the standings so that would be like we would take a schedule and add games to it for no reasons and gave an advantage to a team.

[uruguru]

Originally Posted by AESP_pres

Yes and no... Yes we changed the result of the games (even playing with real transaction / real lineup like I do) but... they were played games, tied games don't count as played in the standings so that would be like we would take a schedule and add games to it for no reasons and gave an advantage to a team.

I am sincerely trying to follow your argument, so please indulge me. What sort of advantage would you be giving a team to change their schedule from 160 to 161 games?


There's enough variability in the replay where it is possible that a team that didn't get a reschedule for a tie or rainout in the real world might actually contend in a simulated replay so that the missing game puts them at a disadvantage.

[AESP_pres]

I guess you will have to think more since it's quite obvious...

[RyMACtizzle]

I think what the OP is saying is that the game was actually on the schedule in 1962. So, theoretically it should be on the 1962 in-game schedule. I don't think he cares too much about your reasoning about what the standings said. The stats counted for the players, so players had the opportunity to play in 161 games or 162 for the 45's despite the standings not showing that amount of games.

So it is not obvious why the in-game schedule doesn't show the game.

[joefromchicago]

Originally Posted by RyMACtizzle

I think what the OP is saying is that the game was actually on the schedule in 1962. So, theoretically it should be on the 1962 in-game schedule. I don't think he cares too much about your reasoning about what the standings said. The stats counted for the players, so players had the opportunity to play in 161 games or 162 for the 45's despite the standings not showing that amount of games.

So it is not obvious why the in-game schedule doesn't show the game.

If the OP is playing the original schedule (the one that the league agreed on and published in the pre-season), then all the teams should be playing 162 games. I understand the OP's complaint to refer to the as-played schedule, not the as-scheduled schedule. Presumably, the as-played schedules don't have games that are rainouts or other cancellations, but they would have the make-up games. Some rainouts aren't replayed, so some teams wouldn't play a full schedule, but if the as-played schedules had both the rainouts and the make-up games, teams would potentially be playing more than 162 games in a season. And really, if the choice is between some teams playing fewer than 162 games to a decision or some teams playing more than 162 games, I think the right choice is the former. After all, that's what ended up happening in real life, so I don't think there's much reason to complain.

I'll add that I don't like how OOTP handles cancelled games for the pre-lights era (roughly before 1940). Today, if a game is halted because of the weather, it is replayed from the point of the stoppage. That's why you almost never see tied games in MLB today. Before the widespread use of lights, however, games that were halted after the fifth inning because of weather or darkness and where neither team qualified for a win were simply declared ties and the game was replayed from the start at a later date. That can't happen in OOTP, which uses the modern approach for all eras.

[David Watts]

Originally Posted by joefromchicago

If the OP is playing the original schedule (the one that the league agreed on and published in the pre-season), then all the teams should be playing 162 games. I understand the OP's complaint to refer to the as-played schedule, not the as-scheduled schedule. Presumably, the as-played schedules don't have games that are rainouts or other cancellations, but they would have the make-up games. Some rainouts aren't replayed, so some teams wouldn't play a full schedule, but if the as-played schedules had both the rainouts and the make-up games, teams would potentially be playing more than 162 games in a season. And really, if the choice is between some teams playing fewer than 162 games to a decision or some teams playing more than 162 games, I think the right choice is the former. After all, that's what ended up happening in real life, so I don't think there's much reason to complain.

I'll add that I don't like how OOTP handles cancelled games for the pre-lights era (roughly before 1940). Today, if a game is halted because of the weather, it is replayed from the point of the stoppage. That's why you almost never see tied games in MLB today. Before the widespread use of lights, however, games that were halted after the fifth inning because of weather or darkness and where neither team qualified for a win were simply declared ties and the game was replayed from the start at a later date. That can't happen in OOTP, which uses the modern approach for all eras.

Just curious, if one doesn't exactly care about everything being 100% true to history, couldn't they simply load the "as scheduled" schedule from the file prior to each season? I'm sure there will be a lot of games that just get OOTP generated lineups and pitchers and one might end up with games with the same pitcher starting back to back games, but it seems like the majority of the games would get "as played" lineups. Or, am I missing something obvious?

[joefromchicago]

Originally Posted by David Watts

Just curious, if one doesn't exactly care about everything being 100% true to history, couldn't they simply load the "as scheduled" schedule from the file prior to each season? I'm sure there will be a lot of games that just get OOTP generated lineups and pitchers and one might end up with games with the same pitcher starting back to back games, but it seems like the majority of the games would get "as played" lineups. Or, am I missing something obvious?

Yeah, you're missing something obvious

I reckon that the OP wants to play with the as-played schedule because he's using the real lineups. I believe that you can't use the as-scheduled schedule if you want to use real lineups, so it's a choice between one or the other - you can't have both. And if you choose the as-scheduled schedule, then the game's AI will determine the lineups - and I don't think the AI's lineups match up very closely with the real-life lineups, especially in earlier eras before sabermetrics and analytics took over. It's also a function of the fact that the AI can look at a player's season stats and abilities before the season even begins, something that managers, of course, can't do.

As near as I figure it, we're dealing with four types of games here:

1. Tie games that were replayed;
2. Tie games that weren't replayed;
3. Cancelled games that were rescheduled and played;
4. Cancelled games that weren't rescheduled or replayed.

The OP is focusing on the second category - tie games that weren't replayed. And I suppose it would be possible to draft the as-played schedules so that those games could be played. After all, they were official games and there were lineups and pitching match-ups, so what's the problem, right? Well, my guess is that no one wanted to go through the schedules and determine which games were in category 2 and which games were in the other three categories. I imagine it would be much easier just to look at all the games that were played to a decision when compiling these as-played schedules. And maybe nobody thought anyone would care if a team played 161 games because there was a tie that wasn't replayed as opposed to a game that was cancelled and wasn't replayed. Not sure. I didn't have a hand in drafting the as-played schedules and I've never done one, so I don't know what the thinking is behind these schedules.

[uruguru]

Originally Posted by joefromchicago

The OP is focusing on the second category - tie games that weren't replayed.

I appreciate someone reading my post and seeing the point as opposed to comically suggesting that I was the one that was confused. And your points preceding that quoted comment were spot-on to what I was thinking.

Look, I get it. Tied games that were rescheduled should NOT be on the actual-played schedule. Tied games that were rescheduled in a different way for one of the teams (e.g. a different opponent to get them to 162 games), should NOT be replayed.

But tied games that were not rescheduled for either team? They should definitely be in the actual-played schedule. They have a starting lineup and pitchers, and their stats counted in the real world. If it's hard to figure it out, I'll spend an hour or two coming up with the list. That's less time than I've spent looking up home towns for players.

And can you imagine your team overperforming to a degree to where, unlike the real world, that missing game actually kept them out of the playoffs? It would be unlikely, but still stupid. But sometimes, even with real world lineups and transactions, things don't play out like in the real world.

[David Watts]

Originally Posted by joefromchicago

Yeah, you're missing something obvious

I reckon that the OP wants to play with the as-played schedule because he's using the real lineups. I believe that you can't use the as-scheduled schedule if you want to use real lineups, so it's a choice between one or the other - you can't have both. And if you choose the as-scheduled schedule, then the game's AI will determine the lineups - and I don't think the AI's lineups match up very closely with the real-life lineups, especially in earlier eras before sabermetrics and analytics took over. It's also a function of the fact that the AI can look at a player's season stats and abilities before the season even begins, something that managers, of course, can't do.

As near as I figure it, we're dealing with four types of games here:

1. Tie games that were replayed;
2. Tie games that weren't replayed;
3. Cancelled games that were rescheduled and played;
4. Cancelled games that weren't rescheduled or replayed.

The OP is focusing on the second category - tie games that weren't replayed. And I suppose it would be possible to draft the as-played schedules so that those games could be played. After all, they were official games and there were lineups and pitching match-ups, so what's the problem, right? Well, my guess is that no one wanted to go through the schedules and determine which games were in category 2 and which games were in the other three categories. I imagine it would be much easier just to look at all the games that were played to a decision when compiling these as-played schedules. And maybe nobody thought anyone would care if a team played 161 games because there was a tie that wasn't replayed as opposed to a game that was cancelled and wasn't replayed. Not sure. I didn't have a hand in drafting the as-played schedules and I've never done one, so I don't know what the thinking is behind these schedules.

Just for fun I gave it a shot. You can import the "as scheduled" schedule when using real lineups and the game will play out the season. Created 1926 and ran it through 1928 importing the "as scheduled" prior to each season. All teams finished the season with 154 games played for each season. Teams started as played lineups and pitchers for the games that fell on the real life scheduled days. On days appearing on the "as scheduled" that differed from the "as played" the OOTP AI created the lineup and selected the starting pitcher. This of course, led to the occasional starting pitcher starting two games in a row.

I see what you mean though and you're correct in regards to what the OP is talking about. Glad I ran the test though, as if I were to use real lineups, I would probably go the as scheduled route. For me it would suck to get to the end of the season with a team that only played 152 games leading by a game.

[joefromchicago]

Originally Posted by David Watts

Just for fun I gave it a shot. You can import the "as scheduled" schedule when using real lineups and the game will play out the season. Created 1926 and ran it through 1928 importing the "as scheduled" prior to each season. All teams finished the season with 154 games played for each season. Teams started as played lineups and pitchers for the games that fell on the real life scheduled days. On days appearing on the "as scheduled" that differed from the "as played" the OOTP AI created the lineup and selected the starting pitcher. This of course, led to the occasional starting pitcher starting two games in a row.

Well, I can file that under "learn something new every day." I didn't know that's how OOTP handled things - which gives some idea of how often I play with real lineups (practically never). Still, your experiment gave me an idea for my own experiment.

As I mentioned, MLB rarely has tied games. In fact, the last time I think there was a tied game was in 2016 (please correct me if I'm wrong). It was a game at the end of the season between the Cubs and the Pirates that was called on account of rain in the top of the sixth inning with the score tied 1-1. The commissioner's office decided not to replay it because that was the last game scheduled between the two teams and its outcome wouldn't have affected the playoff races, so it ended up as a rare tie. How does OOTP handle it? Well, perhaps surprisingly, the game is included in the as-played schedule for 2016 on the date that it was played.

So what does that mean? Well, if uruguru is correct that earlier seasons don't include tied games that were never replayed, then that practice changed at some point, because it now seems that they are included. Why the change? Who knows. Maybe it's as simple as different people having different ways of compiling the as-played schedules. Or maybe tied games are so rare that the person making the 2016 schedule either didn't notice or didn't care that one game ended up as a tie. As I said, I'm not that guy, so I can only speculate.

I see what you mean though and you're correct in regards to what the OP is talking about. Glad I ran the test though, as if I were to use real lineups, I would probably go the as scheduled route. For me it would suck to get to the end of the season with a team that only played 152 games leading by a game.

It happened.

[AESP_pres]

Originally Posted by David Watts

For me it would suck to get to the end of the season with a team that only played 152 games leading by a game.

It happened in my last simulation when the Colonels won the AA championship by half a game over the Brown Stockings in the 1880's because they played one more game.