How do you get your park factors to do what you want them to do?

[Furious]

The manual says the following:

The modifiers are not straight percentages. So, a 2.000 doesn't mean you will do "twice as well."

What would it mean then? If they're not "straight percentages", what are they?

Say I'm trying to get my park to raise batting average for LHB by 3%. What number would I want to input to get that 3%?

[NoOne]

was that park facotrs or league total modifiers?

park factors augment the existing environment. if the manual said that, it just means it's not a straight line on the graph as far as how you augment it through those factors... not %'s to begin with (if that's what the manual said)

sounds more like a section on modifiers.

LTM - league total modifiers from now on

manipulating LTM is intuitive once you look at it the right way. example:

Current LTM - .956 -- current league total 2000

current league resuts from simulations ~1550hr (over the long term, not 1 year)

intended results: ~2000ish ** whether LT or not - "what you want"

(**doesn't have to be the total. if you want something different it's best to change the modifier and leave the league total alone -- only drawback is any time you want to click auto-calc you'll hve to re-adjust anything you did before that is not set for "2000" -- same exact process so no big deal)

so 2000/1555 = 1.286 --- a ~28.6% increase needed. now go back to the original LTM used:

.956 * 1.286 = 1.230 new HR LTM

the % change is not relative to "1" or "1.000", it is relative to the current state. all changes occur from what you are using "now"

------------


If the LT is not the # you want and you have recently hit auto-calc:

let's say "2000" in the example above is not the actually LT. the LT is "1555" for simplicity here inthe example.

So, each time you click auto-calc, you'll have to add 12.86% (2000/1555 in this case -- wanted #/current #) to HR LTM after auto-calcing at any time in the future to get roughly 2000 average over long-term

(remember not to focus on 1 year's results... if it fits teh range, it's potentially okay... for HR's it's ~10% sway up/down from average in any given year of ootp*** that sway may increase or decrease if the total is larger/smaller, respectively not 100% sure on that but pretty sure i noticed it in the past)


The complicated part:

you change one stat it likely affects others. e.g. HR will reduce BABIP. if you like your current babip and want to keep it after the hr ltm change, you'll need to adjust that too.

hits also affects doubles, triples etc... oyu increase hits 10% that will be divied up amongst all "hit types" in the ratio that the current LTM/LTs amount to in simulation over the long term. (hit type? = what constitutes a hit in baseball etc)

you can see how it dominoes. small changes i wouldn't worry too much... maketh adjustment you want, then minor adjustments in ensuing years if needed to other things.

big adjustments like the example above 1500hr to 2000 home runs, you may want to try to work out the domino or use some trial and error simulations with a restored backup copy of your league.

not a suggestion for the weak-at-heart.... i have a spreadsheet in forums that helps set specific LTMs and even your LTs if you want to do that... the only trick with changing LTs is making sure they add up properly...

e.g. if you randomly choose numbers and you hr/bb/hits wont add up to a proper BABIP as designated in your LT. this will likely cause odd results or at least unintended results. the spreadhseet basically calculates teh totals that need to be calculated from other totals and you change the ones that control those...equals a simply made LT set

[geoffmatt17]

No, OP is definitely talking about park factors:

Ballpark Factors
The Ballpark Factors section directly impacts how hitters perform in the selected ballpark, over a large sample size. Ballpark Factors are extremely important, as they directly affect the outcome of plays in a given park.

Ballpark factors are based on a "norm" of 1.000. That is, a ballpark with all 1.000 factors is essentially a "neutral" park where hitters will all perform similarly. As the numbers increase, that factor becomes more common. So, for example, if your AVG Overall factor is 1.100, you can expect that if you had identical players in this park and a neutral park, the player in the park with the 1.100 AVG Overall factor would have a slightly higher average. The modifiers are not straight percentages. So, a 2.000 doesn't mean you will do "twice as well."

http://manuals.ootpdevelopments.com/..._page.overview

[Rain King]

So, there is no guarantee that the league average of your park factors is 1.000. If the average modifier for your leagues BAVG is 1.2, then 2.0 is not going to mean double. Your stats are still first controlled by the League Totals.

[Furious]

Not really looking for a guarantee. More looking for a—excuse the pun—ballpark figure.

Let's say, for the sake of argument, that the league average of all my ballparks is 1.000. Then would a park factor of 2.000 mean "twice as well"?

We're having a discussion in our league on whether to put limitations on what park factors we allow, and what those limitations should be. We're stuck right now because no one really knows what kind of an effect these park factor values have. Do you need a really large number to have a noticeable effect? Or will a smaller number do?

I understand that league totals modifiers effect everything, but it's irrelevant to the discussion. We're not trying to get exactly X amount of home runs hit in a particular park. We're trying to figure out what are reasonable values to use for park factors. We're not asking for a "this will produce exactly this number of home runs" answer, but we do need something more precise than "2.000 doesn't mean you will do twice as well."

[conception]

For a good idea of what park factors should be perhaps the best way is to just look at park factors for real parks either on the internet or just in the game.

[Furious]

Originally Posted by conception

For a good idea of what park factors should be perhaps the best way is to just look at park factors for real parks either on the internet or just in the game.

I completely agree with that, but some people in our league believe you can enter much higher values without it having much of an effect. Without word from someone who really knows how the park effects work I can't say they're not right.

[conception]

Perhaps the easiest solution then would be to have people pick out the park factors of real life ballparks as their in-game parks. Asking a whole group of online owners to understand how park factors work and be honest with them is too much.

[Sweed]

Originally Posted by Furious

Let's say, for the sake of argument, that the league average of all my ballparks is 1.000. [I]

Well FWIW this is the way it is supposed to be setup. IE the total of all factors, for every stat, should average 1.000. From the manual

Ballpark factors are based on a "norm" of 1.000.

So for every owner in your league that wants to increase a stat by .3 another has to decrease by that much or you lose your "norm" of 1.000.
If you have guys just making these up "on the fly" good luck on having any idea of what your stat output will look like.

[RonCo]

Take this for what it's worth...every league will be a little different for many possible reasons, but here are some coarse data...

1.I created at simple test league of 8 teams on pretty much all default settings.
2) did a roster export, and did a very quick adjustment of all main ratings to "100", then imported. I did not adjust pitcher repertoire, so there's noise in that data.
3) Set all ballpark factors of each team to a single setting (Team #1 = 2.00, Team #2 = 1.6, etc).
4) Then I ran one year of data.

Here are the results:

[Furious]

Originally Posted by RonCo

Take this for what it's worth...every league will be a little different for many possible reasons, but here are some coarse data...

1.I created at simple test league of 8 teams on pretty much all default settings.
2) did a roster export, and did a very quick adjustment of all main ratings to "100", then imported. I did not adjust pitcher repertoire, so there's noise in that data.
3) Set all ballpark factors of each team to a single setting (Team #1 = 2.00, Team #2 = 1.6, etc).
4) Then I ran one year of data.

Here are the results:

Clear as a bell. Thank you.

[USBL Commissioner]

I'm sorry if this is the wrong thread to ask this question, but I see you are discussing ballpark factors/modifiers...and was hoping if anybody can answer a question for me. I have created a fictional league (Arena Baseball League) and have modified both the league settings as well as the ballparks. The regular season is great....but spring training is played in a "basic" park. Is there any way to update the factors for spring training?

[Drstrangelove]

Park factors are basically a calculation of the park's effect on the runs scored in a park relative to the league while normalizing for teams. For example, when a park factor is 1.10, it means that games in that park are expected to average 10% more runs than the league average assuming the teams are average teams. So it's adjusting for league averages and normalizing all the teams and inferring the effect of the park.

At the same time, runs created (which is a calculated derivative of how actual runs are scored) is based upon the expected results of actual batting performances, such as singles, homers, walks, errors, stolen bases, etc.

So while a park factor of 2.000 implies an average team will score 100% more runs in that park than an average park, it doesn't mean that all the contributing metrics (singles, homers, walks, errors, stolen bases, etc.) will double. E.g., a team that gets 18 hits and 4 homers in a game on average will likely score more than twice as many runs as a team with 9 hits and 2 homers. The reason for this is obvious. The more hits you get, the more likely you will bunch them with runners on base. So, the expected value of a single changes under different contexts. So that's partly what the "twice as well" comment means. The other part is of course because a team only plays one half of it's games at home.

The problem as I see it (and the simulation RonCo posted shows that problem) is that when you create an aberrant metric, you'll get aberrant results. In the simulation, the average park factor for the league was around 1, yet the league was scoring 7.5 runs per game. That's likely because of using aberrant park factors like 2.000. If you let 3 teams do that, you may average 40-50 runs per game. The results are unpredictable.

In that sim, there were probably dozens of games at that park (and only at that park) where a team scored over 50 runs because pitchers were used up and one or more pitchers was dead tired and left in the game to give up 20-30 runs. That's just a guess. But it makes sense since a team playing in a park with a .40 park factor still managed to score 750 runs. So letting teams set park factors to 2.000 will distort the league.

This will likely end up grossly distorting scouting results, salary demands, free agent signings, draft choices, etc. Games will be much longer with a lot more pitches, which will likely lead to more pitching injuries especially if you use true injuries settings.

If it were my league, I would limit teams from setting park factors above 1.20.

[Furious]

Originally Posted by Drstrangelove

If it were my league, I would limit teams from setting park factors above 1.20.

The issue is that there was disagreement in our league on how much park factors affected things, largely due to the vague language in the manual, and also some information in some other threads (which may have been somewhat out of context). I was hoping someone on the programming side of OOTP would chime in and make some definitive statements to clear things up, because what the discussion was missing most was informed opinions.

Even with the flaws you pointed out, RonCo's study at least showed that a park factor of 2.000 is an extreme, and that extreme factors give extreme results. I really didn't care if 2.000 meant exactly 200%; I needed to know if it meant something closer to 2%, or 2000%. We now have more information than we had at the start.

[RonCo]

Originally Posted by Drstrangelove

Park factors are basically a calculation of the park's effect on the runs scored in a park relative to the league while normalizing for teams. For example, when a park factor is 1.10, it means that games in that park are expected to average 10% more runs than the league average assuming the teams are average teams. So it's adjusting for league averages and normalizing all the teams and inferring the effect of the park.

This is not what OOTP park factors do.

The "problem" (Assuming there is one. ) is that people bring reasonable expectations like you are bringing into the game, and the game uses language that makes that feel right. But that's not what the game itself does.