June Release

[swampdragon]

Originally Posted by PSUColonel

They are apples and oranges. OOTP will never be able to do both well. It is not a replay sim, it is a GM management game.

Don't count on that. The in-game engine has nothing to do with the GM, and nothing has to be sacrificed on the GM or fictional side to improve it. I think Markus can make this the perfect game. There's no really good reason it can't be both. As Syd pointed out, the way recalc is handled and the use of DIPS introduce more realistic variability than DMB can provide. I truly believe that if you don't particularly care about playing the games out, OOTP will be at least as good as DMB this year. Plus it's going to keep moving forward, and DMB isn't.

[Russ]

I just wish the PbP was better. It's improving, but it still needs more news involved rather than just a grounder to short and the batter's retired.

Ex.

"Calderon is stepping up to the plate with a 25-game hitting streak on the line."

"With that home run, Smith is now third on the all-time list."

etc, etc.

[endgame]

Originally Posted by Russ

I just wish the PbP was better. It's improving, but it still needs more news involved rather than just a grounder to short and the batter's retired.

Ex.

"Calderon is stepping up to the plate with a 25-game hitting streak on the line."

"With that home run, Smith is now third on the all-time list."

etc, etc.

While I am confident that there are always changes at work in pbp - probably always will be - IMO, a premiere addition in some distant future version would be the ability to access leaderboards, league histories, etc. while in-game; providing everything outside the game, inside. It'd be so much easier to answer those quick in-your-head color commentary questions we all have during the course of play.

[Le Grande Orange]

Originally Posted by Charlie Hough

By putting Anderson on the roster as one of the top relief pitchers from the start of the season, he will end up pitching far more games and innings than he did in real life. It would be unrealistic.

The problem there is how far do you want to follow that logic?

One could say that if a real-world player missed a lot of games due to an injury in a given season, since he likely won't experience that injury in OOTP, he'll get more playing time than he did in real life. By your definition, that's unrealistic.

One could say that if a player was traded at a certain point in the season, and OOTP doesn't recreate that, then it's unrealistic since the player will have more games with the other team than he did in real life.

And so on.

So where does the realism line get drawn?

[trav76]

Can someone give me a quick idiots guide to historical play? Do players still develop like a normal game, do you trade that sort of thing? Or are you bound to where players played in real life? Can a player that wasn't so crash hot in real life be developed into a class player? And alernatively can a class player be hampered by injuries, poor development etc?

[emcgman]

When you create a historical league, you can check a box to have the game recalculate historical ratings/stats for every year, or if left unchecked the players will develop randomly, according to their raw talents.

So you could play it either way.

[emcgman]

Hey Russ,

I really like your examples of PbP.

Man, if they could put that kind of stuff in, PbP would be on a whole other level.

[Gerry58]

Originally Posted by trav76

Can someone give me a quick idiots guide to historical play? Do players still develop like a normal game, do you trade that sort of thing? Or are you bound to where players played in real life? Can a player that wasn't so crash hot in real life be developed into a class player? And alernatively can a class player be hampered by injuries, poor development etc?

Historical replays with OOTP are great. You can import whatever season you like (i used the Lahman database) and start. Yes, the players develop and there are injuries (but you can turn them off) and you are NOT bound to follow real life trades. That's an area that needs improvement (the trade AI is not very smart) but i'm confident that the next version will be better. As an example, i started a 2002 historical replay with the Yankees. You start in the off-season so you can prepare for trades before spring training begins: i traded a lot but the end result was not so good: lost to the Indians 2-3 in the first round of the playoffs. I can't wait for the next version.

Gerry

[ctorg]

Originally Posted by Syd Thrift

I'm already at a point where I can't even look at DMB because of realism issues. Sure, base advancement may be based on close readings of Retrosheet, but there's no DIPS. Unless you are playing seasons through exactly as in real life, there will be differences in realism. Is Jim Palmer in the 70s half as good as IRL if you traded him to Cleveland? Probably not in a realistic universe because he had a low lifetime BABIP due to playing in front of Brooks Robinson, Mark Belanger, and Al Bumbry and that would surely rise if placed in front of Tom Veryzer and Rick Manning (Buddy Bell was pretty decent, at least). In DMB, you can expect him to play pretty similarly - from my observations, a bad defender affects a high-strikeout pitcher pretty much exactly as much as he affects a low-K guy (I know this to be the case with SOM because I played that game for years before turning to OOTP).

Also, to me "fog of war" means so much. Why did the Braves give 81 at-bats to Matt Sinatro in 1982? Surely they did not realize that he would hit .136 that season. And yet, in a game like DMB you're either forced to play a .136 hitter or "cheat" by giving his at-bats away to someone else. I find it much, much more realistic to not really know how awful a Matt Sinatro type until I give him at-bats (I can scout him, but what if my scouting is inaccurate?).

To add to that, I doubt Sinatro was "really" a .136 hitter that season. He was probably more like a .200 hitter who got into a bad slump. Granted, I wouldn't want to waste at-bats on a .200 hitter with no sock and little patience either, but that meme expresses itself in other ways. You replay the 1961 season with DMB or SOM, Roger Maris gets 61 or more homeruns roughly half the time. That may seem realistic to some, but to me, reviewing a career of a man who never otherwise got above 40 dingers in a career, I think he should only break Ruth's record 10, 20 percent of the time. It was a fluke and ought to be represented as such. You simply can't do that in a single-season replay and maintain a shred of accuracy.

Well, if your concern is accurate recreation of a season rather than accurate representation of players according to their ability, then BABIP and DIPS stuff might not be relevant. After all, if you want pitchers to put up similar stat lines to real life, you'll want them to have a similar BABIP to real life, too.

That's why I prefer OOTP to DMB or SOM or other games. It gets more into what makes the players perform well rather than just trying to get them to put up similar numbers.

[Charlie Hough]

Originally Posted by Le Grande Orange

So where does the realism line get drawn?

I'm glad to asked, because I should probably clarify my own arbitrary definition of what constitutes realism.

My preference for 'replay' sims, dating all the way back to when I started, was to have something that would accurately reflect the ability of regular players and make sure that marginal players cannot be abused.

I'm not concerned with duplicating stats, but I would expect statistics to be within a reasonable range of real life. I can't be very exact, but, for example, I wouldn't want to see a .320 hitter hit anything better than .340 or worse than .300. Somewhere in that range would be reasonable. Likewise, I would not want to see a player with 15 home runs hit well under 10 or more than 25. This all assumes roughly the same numbers of at bats and similar usage overall.

The main thing is to recreate the basic conditions of the teams and the season and then see how the games unfold. Those basic conditions would include a proper rendering of player ability and a statistical integrity that does not stray too far from real life.

OOTP captures that, but the one area where it has been lacking is in the realism of roster selections, lineups, and pitching staffs. It doesn't seem to limit those marginal players enough, so that's how you can end up with a utility player becoming a regular and ending up with 500% of his actual usage.

However, I have to make it clear that my interest in 'replay' sims has always been to see how things play out differently. It's not to see how close to the actual stats that the sim can get. If I want the actual stats, I can just pull out my Baseball Encyclopedia.

For me, the fun has always been to take a bad or mediocre team and see if I can get it to achieve better results by using my own management style and sometimes making some fair and realistic trades that give me the kind of team that I want. It's also fun to see if the AI clubs do the same, better, or worse than they did in real life.

So, for me, it's not an issue if Ed Lynch ends up pitching a lot of games for the 1986 Mets. The conditions at the start of the season included Lynch on the pitching staff. Rick Anderson was nowhere to be found. Now, a game like Strat-O-Matic will make sure that Ed Lynch gets injured like he did in real life, at least at some point in the season, and he'll be limited in most sims to about the same usage he enjoyed in real life. If you truly want exact usage, you can actually force all the changes that happened in real life. But, personally, I like to have Ed Lynch on the Mets' staff, because he was a far better pitcher than Rick Anderson in real life, and to me it's more consistent with the conditions at the start of 1986.

This is why I prefer OOTP and why I have not played other sims for years. OOTP reminds me a lot of MicroLeague Baseball in that it provided a good rendering of player ability and statistical integrity without being so tied to real life statistics that you couldn't make much of a difference as a human manager. With the new improvements for OOTP 9, I think the game will have the best balance that we've ever seen between hardcore statistical realism and the fun of seeing how a simulation might deviate from real life in some interesting but plausible ways.

[StyxNCa]

Originally Posted by Charlie Hough

I'm glad to asked, because I should probably clarify my own arbitrary definition of what constitutes realism.

My preference for 'replay' sims, dating all the way back to when I started, was to have something that would accurately reflect the ability of regular players and make sure that marginal players cannot be abused.

I'm not concerned with duplicating stats, but I would expect statistics to be within a reasonable range of real life. I can't be very exact, but, for example, I wouldn't want to see a .320 hitter hit anything better than .340 or worse than .300. Somewhere in that range would be reasonable. Likewise, I would not want to see a player with 15 home runs hit well under 10 or more than 25. This all assumes roughly the same numbers of at bats and similar usage overall.

The main thing is to recreate the basic conditions of the teams and the season and then see how the games unfold. Those basic conditions would include a proper rendering of player ability and a statistical integrity that does not stray too far from real life.

OOTP captures that, but the one area where it has been lacking is in the realism of roster selections, lineups, and pitching staffs. It doesn't seem to limit those marginal players enough, so that's how you can end up with a utility player becoming a regular and ending up with 500% of his actual usage.

However, I have to make it clear that my interest in 'replay' sims has always been to see how things play out differently. It's not to see how close to the actual stats that the sim can get. If I want the actual stats, I can just pull out my Baseball Encyclopedia.

For me, the fun has always been to take a bad or mediocre team and see if I can get it to achieve better results by using my own management style and sometimes making some fair and realistic trades that give me the kind of team that I want. It's also fun to see if the AI clubs do the same, better, or worse than they did in real life.

So, for me, it's not an issue if Ed Lynch ends up pitching a lot of games for the 1986 Mets. The conditions at the start of the season included Lynch on the pitching staff. Rick Anderson was nowhere to be found. Now, a game like Strat-O-Matic will make sure that Ed Lynch gets injured like he did in real life, at least at some point in the season, and he'll be limited in most sims to about the same usage he enjoyed in real life. If you truly want exact usage, you can actually force all the changes that happened in real life. But, personally, I like to have Ed Lynch on the Mets' staff, because he was a far better pitcher than Rick Anderson in real life, and to me it's more consistent with the conditions at the start of 1986.

This is why I prefer OOTP and why I have not played other sims for years. OOTP reminds me a lot of MicroLeague Baseball in that it provided a good rendering of player ability and statistical integrity without being so tied to real life statistics that you couldn't make much of a difference as a human manager. With the new improvements for OOTP 9, I think the game will have the best balance that we've ever seen between hardcore statistical realism and the fun of seeing how a simulation might deviate from real life in some interesting but plausible ways.

This is word for word how I was going to answer.

[holley]

[quote=Charlie Hough;2484086]The book is the Team by Team Encyclopedia of Major League Baseball by Dennis Purdy. It contains the primary starting lineups and pitching staffs for all MLB clubs from 1876-2005. Unfortunately, it does not provide the actual batting orders, probably because those vary so much compared to the positional lineups and couldn't be easily provided in a book.

Hey Charlie, Just wanted to drop a note to say thanks for the tip on the book. I picked it up thru Amazon for a $1.16, $3.95 shipping. I thought you might appreciate the "good" deal I got. Anyways, love the book. Thanks so much for the tip.

Cheers.

M.

[FleetWalker]

Originally Posted by Charlie Hough

I'm not concerned with duplicating stats, but I would expect statistics to be within a reasonable range of real life. I can't be very exact, but, for example, I wouldn't want to see a .320 hitter hit anything better than .340 or worse than .300. Somewhere in that range would be reasonable. Likewise, I would not want to see a player with 15 home runs hit well under 10 or more than 25. This all assumes roughly the same numbers of at bats and similar usage overall.

That might be a slightly aggressive definition of realism - at least mathematically.

I did a quick back of the envelope calc, and assuming the hitter has the same .320 chance to get a hit in each AB, and that there are 525 ABs in a season, then 90% of the time the hitter will have a batting average between .354 and .288. That may sound OK, but it also means that 10% of the time a "true" .320 hitter will hit more than .354 or less than .288 just because of random chance. Indeed, 1% of the time, the batter will hit more than .373 or less than .269.

The only way to "fix" this is to adjust a hitters chance of being successful based on how well they are doing. Of course, if you do that you'll mess things up for pitchers.

[jetking]

Originally Posted by Charlie Hough

Unfortunately, it does not provide the actual batting orders, probably because those vary so much compared to the positional lineups and couldn't be easily provided in a book.

For batting orders (back to 1958, at least) I recommend Baseball-Reference.com. They show game-by-game as well as the six most common orders for the team for that year.

Some examples:

1958 Dodgers, 1963 Angels, 1977 Rangers, 2005 Astros

I hope that helps someone.

[Curtis]

Originally Posted by Syd Thrift

…from my observations, a bad defender affects a high-strikeout pitcher pretty much exactly as much as he affects a low-K guy (I know this to be the case with SOM because I played that game for years before turning to OOTP).

I also played SOM for years, though not in the past four years. My experience was more with the good fielders than the bad. In any given season my Dodgers would have five to seven Gold Glovers on the field. Between that and Dodger Stadium being a pitchers' park with very wide foul ground, my pitchers tended to do much better for me than they did historically.

Originally Posted by ctorg

That's why I prefer OOTP to DMB or SOM or other games. It gets more into what makes the players perform well rather than just trying to get them to put up similar numbers.

Sometimes the numbers aren't all that similar. The manager who had Mark McGuire during his heyday never got 60 homers from him, and neither did the manager who had Sammy Sosa. Few people stole as many bases in the game as they did in real life because we tended to make our catchers with the good arms our everyday catchers, and leave the good hitters for injury/late inning/pinch hitter/fatigue relief purposes.

On the flip side of that, the manager who had Barry Bonds was in clover. That guy regularly hit more out in our game than he did in real life, AND he drew more walks. One of the major shortcomings of that game engine was that it counted real life intentional walks the same as any other. Thus, Barry would draw 100+ IPP in real life, and that would make our pitchers have to give him what were in effect intentional passes in stupid situations in our game. Add to that the real intentionals we fed him in important game situations, and he set records that should never be approached.

Originally Posted by Charlie Hough

Now, a game like Strat-O-Matic will make sure that Ed Lynch gets injured like he did in real life, at least at some point in the season, and he'll be limited in most sims to about the same usage he enjoyed in real life.

Not in the versions I played. There WERE three settings (ignore real life usage, enforce real life usage, try to get close to real life usage). Since did a lot of trading, enforcing real life usage might result in having to forfeit the last fourteen games of the season due to an insufficient number of players being available. We opted for trying to get close to real life usuage, but it wasn't always real close.

For the 1997 season, Mike Piazza went down with a 15 day injury on May 1st and never returned because the computer GM was too stupid to bring him off the disabled list. (We simmed a season in fifteen minutes.) Instead it started his backup, who played two games in real life, for over 130 games, and that team missed the playoffs by one game.

On my own club two relievers who totaled 20 innings in real life combined for 200 in the game. Since they both had ERAs below 1.00, I was a happy, happy man, but all of the managers were disgusted enough that we put some safeguards in place to prevent that from happening again.

On a non-SOM note to Charlie, why is the situation at the start of the season so much more important to you than (for example) the situation at the All Star break, the trading deadline, or September 1st?

[Mariner and Giants Fan]

Originally Posted by Markus Heinsohn

If you have 3-year recalc enabled, which is the default, then the results are no wonder. Try 1 year recalc, that will get you the results you prefer.

I just imported 1986 into OOTP 9, using 3 year-recalc, and the Mets closer is Randy Myers, which makes sense since he has the best ratings... you need to check in the player editor, you see the expected ERA there which is computed using the ratings. Myers has an expected ERA of 2.92 (in neutral enviroments), and Orosco 3.47 ... so, the AI in OOTP 9 did exactly the right thing, and if you check the K/BF and K/BB numbers, you see that Myers was the better pitcher stats-wise in that period as well.

Edit:
Checked the 86 import with 1-year recalc, and this time the closer is Anderson, who indeed has the best expected ERA. Orosco was pretty lucky in real life in 86, with a .254 BABIP, a BB/9 of almost 4 and just a 6.89 K/9. Again, the AI does the right thing.

The screen to do the recalcs? I am having the same problems with historical AI.

Thanks.

[Charlie Hough]

Originally Posted by Curtis

On a non-SOM note to Charlie, why is the situation at the start of the season so much more important to you than (for example) the situation at the All Star break, the trading deadline, or September 1st?

Because, if you have the wrong guys in the lineups and on the pitching staff at the start of the season, then you run into the kind of situation like Curtis mentioned in reference to SOM. A marginal player with only two games played in real life could end up playing 130 and ruining your sim results.

If OOTP could give us a better way to enforce lineups, pitching staffs and roster changes -- at all points of the season -- then I would definitely say that it's just as important to get it right at the other points you mentioned.