91 Home Runs! Should I believe it?

[Cobby]

Clint Sampson had quite the season in 1996. He hit .404 with 91 home runs!

But. I'm not sure if I should believe it or if there is something awry with my setup or the simulation. I don't think it's my setup - I've got player creation modifiers all at 1.0. Recalc is on. It's a fairly small league with 24 teams, reserve clause - no minors. All of the ballparks are set for neutral factors.

I haven't actually been playing this league - just simulating through history to get a feeling for how well the simulation does compared to real baseball history.

The simulation seems homer-happy - especially in more modern times. I've had 1 90+ home run season, 4 80+ home run seasons, 12 70+ home run seasons and fully 48 60+ home run seasons.

Career home runs are high too. Four players have over 800 career home runs, 10 have over 700 and 21 have over 600.

I don't remember seeing this before. In past versions the modern results seemed pretty good - it was early 20th century baseball that seemed off.

So. I'm wondering if anyone else is seeing this. Has something changed in the simulation engine?

One possible clue: each year I've been calculating the league average rating for each of the ratings. I've been doing this by calculating a weighted average by plate appearance. For most of the ratings the league average is consistent and around 50 (on a 1-100 scale, scouting accuracy at 100% of course). For Power though, the league average is still consistent but it's around 36. This is probably intentional, but it means that when a player has a high power rating it'll be further above the league average than the other ratings.

Sampson's power rating maxed out at 84 (1-100) scale, but he still hit those 91 homers. Imagine how many he would have hit with a power rating closer to 100...

Any thoughts?

[Findest2001]

Dang, that dude never gets injured

[monkeyman576]

Kansas City must have been really bad to have lost him after he hit 40 HR/.360/150 some RBI's.

[Reed]

Sounds like your setup might be messed up. I never mess with player/league modifiers.

[Cobby]

Originally Posted by Reed

Sounds like your setup might be messed up. I never mess with player/league modifiers.

That's my point - having player creation modifiers set to 1 is the default. (for the modern era at least) I don't remember if having auto-recalc on is the default or not - but all that does is generate a set of modifiers each year based on the current composition of your league so that the yearly league-wide stats stay close to the historical league-wide stats for that year.

There are some on this forum who say that yearly re-calc tends to dampen extraordinary performances rather than encourage them. But for what I was trying to do - doing a fictional historical simulation in order to see how the results through time match up with real historical results - I definitely needed to recalc along the way. (although probably not necessary to do it every year) (but without recalc, you can't hope to get the transition from the dead-ball to the live-ball era, for instance)

[texasmame]

Originally Posted by Cobby

That's my point - having player creation modifiers set to 1 is the default. (for the modern era at least) I don't remember if having auto-recalc on is the default or not - but all that does is generate a set of modifiers each year based on the current composition of your league so that the yearly league-wide stats stay close to the historical league-wide stats for that year.

There are some on this forum who say that yearly re-calc tends to dampen extraordinary performances rather than encourage them. But for what I was trying to do - doing a fictional historical simulation in order to see how the results through time match up with real historical results - I definitely needed to recalc along the way. (although probably not necessary to do it every year) (but without recalc, you can't hope to get the transition from the dead-ball to the live-ball era, for instance)

If the rest of the league's offense seems to be normal that year, guess he just went wild. His name *is* Sampson, after all!

[Cobby]

Originally Posted by Reed

Sounds like your setup might be messed up. I never mess with player/league modifiers.

However, your post got me thinking. I haven't been checking what the modifiers actually come out to after the recalc, but I just checked for the current season (I have simulated through 2019) and the one modifier that deviates the most from 1 is the home run modifier. That one is currently around 1.5.

In some ways, that makes sense, since home runs in baseball have been up considerably recently. But it also could be at the root of the problem - unlike all of the other ratings the average power rating of my league is has been consistent through the years at around 36. So when the re-calc comes along it sees that based on 2018 stats the league-wide home run total is pretty high, but the average power rating of the players in the league is low so the modifier is set high to compensate. This modifier applies to all the players in the league so that if you happen to be one of the few with a high power rating, you also get the benefit of that 1.5 modifier and might end up hitting 90 home runs.

Anyway, that's what I'm thinking - that maybe something has changed internally with player creation that is causing the power rating to behave differently from the other ratings. I'm thinking that this *may* be the case, but probably unlikely. More likely is that the developers think carefully about these things and decided that there needed to be a larger spread between average and elite home run hitters and adjusted their player creation algorithm accordingly and that's why the average power rating is lower. Nevertheless, maybe something did go wrong here in the latest version...

[Cobby]

Originally Posted by texasmame

If the rest of the league's offense seems to be normal that year, guess he just went wild. His name *is* Sampson, after all!

Ah yes! That's got to be the problem. The game is not generating enough pitchers with the last name of "Delilah"

[Findest2001]

Originally Posted by Cobby

Ah yes! That's got to be the problem. The game is not generating enough pitchers with the last name of "Delilah"

Delilah wouldn't be a pitcher, silly. She'd be his drug dealer selling him some tainted steroids.... Amateurs.

[Rizon]

Originally Posted by Cobby

Clint Sampson had quite the season in 1996. He hit .404 with 91 home runs!

But. I'm not sure if I should believe it or if there is something awry with my setup or the simulation. I don't think it's my setup - I've got player creation modifiers all at 1.0. Recalc is on. It's a fairly small league with 24 teams, reserve clause - no minors. All of the ballparks are set for neutral factors.

I haven't actually been playing this league - just simulating through history to get a feeling for how well the simulation does compared to real baseball history.

The simulation seems homer-happy - especially in more modern times. I've had 1 90+ home run season, 4 80+ home run seasons, 12 70+ home run seasons and fully 48 60+ home run seasons.

Career home runs are high too. Four players have over 800 career home runs, 10 have over 700 and 21 have over 600.

I don't remember seeing this before. In past versions the modern results seemed pretty good - it was early 20th century baseball that seemed off.

So. I'm wondering if anyone else is seeing this. Has something changed in the simulation engine?

One possible clue: each year I've been calculating the league average rating for each of the ratings. I've been doing this by calculating a weighted average by plate appearance. For most of the ratings the league average is consistent and around 50 (on a 1-100 scale, scouting accuracy at 100% of course). For Power though, the league average is still consistent but it's around 36. This is probably intentional, but it means that when a player has a high power rating it'll be further above the league average than the other ratings.

Sampson's power rating maxed out at 84 (1-100) scale, but he still hit those 91 homers. Imagine how many he would have hit with a power rating closer to 100...

Any thoughts?

Can you post your league totals and league modifiers?

[Cobby]

Originally Posted by Rizon

Can you post your league totals and league modifiers?

Sure. The totals ought to be 2018 totals. They had been updating every year until I reached 2019. The modifiers of course will also update each year when the program does the auto-calc.

[Mat]

I would manually drop the HR modifier down to something you are happy with and then untick auto adjust league totals modifiers.

[kq76]

Do you have dynamic evolution on?

Those modifiers are way out of whack. I can't right now, but I would test what a newly created MLB league's totals and modifiers are in OOTP 20.

Also, in case you don't know, and if it hasn't changed, changing totals creates an inverse effect. That is, lowering your home run totals will actually increase them in the future.

[Cobby]

Originally Posted by kq76

Do you have dynamic evolution on?

Those modifiers are way out of whack. I can't right now, but I would test what a newly created MLB league's totals and modifiers are in OOTP 20.

Also, in case you don't know, and if it hasn't changed, changing totals creates an inverse effect. That is, lowering your home run totals will actually increase them in the future.

No, dynamic evolution is off.

The modifiers do seem to be way out-of-whack, but they are the correct modifiers for my current talent pool. I've attached a comparison between the simulation results and real-life 2018 results. The rates are very close. They wouldn't be close without those particular modifiers.

So the problem isn't really the league totals or modifiers, it's why my talent pool requires such out-of-whack modifiers.

I don't know the answer, but here are a few thoughts:

For one thing, we do seem to be entering a new era in baseball with home runs and strikeouts occurring at much higher rates than even just a few years ago. When you have a standard MLB league the ratings of the players can reflect this right off the bat (ratings, being based on projections, will automatically reflect the probabilities of all of the potential events) so the modifiers can be near 1.

On the other hand, for a fictional league you don't get to pre-tailor the ratings - you have to depend upon the player creation algorithm. Then if you want things come out statistically in a particular way, you do the auto-calc to get things to match what you want. And. That seems to work really well at reproducing whatever overall league stats you want. But...

I think it does become a problem for individual performances. I suspect that having a large modifier will expand the difference in results between the top players in the league and the average (just like multiplying a set of numbers by a constant greater than one will increase the standard deviation of that set of numbers).

The program deals with this for most of baseball history by having different player creation modifiers that reflect the changing style of baseball through the years. But these all end up at 1 and stay there starting in 2005 so they aren't tracking the recent changes in style. (check out the era_modifiers.txt file in the database folder)

Another thought though - maybe it's my lack of minor leagues. I have 5 draft rounds and creating players for 6 rounds. This is the default for a no-minors league, but maybe it's messing up the talent pool. I could try increasing the number of rounds or adding minor leagues and see if it makes a difference. I sort of doubt if it would make a difference since I think the developers would have thought about this and would manage the talent pool to get reasonable results for whatever setup you're using. But I could try it and see.

Anyway. This is mainly directed to the developers - they're the ones who know how all of this works internally, so they ought to be able to determine if this is a problem or working as intended. Of particular interest is whether of not the low PA-weighted power rating of 36 is intentional or represents a problem. Just for completeness, here are all of the PA-weighted (or BF for pitchers) league average ratings:

Contact: 47.5
Gap: 57.7
Power: 36.6
Eye: 46.6
Avoid K's: 53.5
Stuff: 63.7
Movement: 41.0
Control: 48.1

In the meantime, there are a few things I can play around with and see if I can get a realistic and exciting alternate baseball history...