All Forums > SimLeague Baseball > SimLeague Baseball > Home Field Advantage
7/15/2012 2:15 AM
Fair warning, this is going to be lengthy.

For a number of years now we've had various discussions about home field advantage here on in the forums with most of them following the same pattern of discussion: An owner asks why they seem to have a better road record when they've tailored their team to their ballpark, a number of owners post anecdotes about the majority of winning teams doing better on the road, an owner will post general data about the visiting team winning more because they score first, I post my current W-L records detailing a clear advantage at home and the topic dies.

I'd never really thought much about it except to assume that those who weren't winning at home were probably just not tailoring their teams as well as they thought they were. Then, the other day, as I was at work looking at scores on my phone I saw my record once again had me doing very well at home, and below .500 on the road, while being one of eight teams with a .550 or better W% in an OL.

Then, just because I was curious, I looked at the other teams home/road splits and noticed that all of the NL teams above .550 were better at home and all of the AL teams were better on the road. A mere curiosity until I saw that one of the AL teams belonged to sford. And then it hit me. The owners that I typically see with better road records tend to build teams around ballparks that are friendlier to their pitchers. So, I did some digging around in that league, other leagues I'm currently in, through all of my past leagues that I've saved data on, and came to some interesting conclusions.

In general, teams that choose a ballpark that is neutral to pitcher friendly will have better road records while teams that choose a ballpark that is more offensive friendly will have better home records. So, obviously, I wondered why that was the case and was instantly drawn by the comments in the older threads on home field advantage and the theory that the team that scores first wins. So, without having hard data on this, I ran with that assumption and came to a quick summary of:

- When in a neutral or pitcher friendly environment, runs are harder to come by, so the team that scores first is more likely to win. This benefits the visiting team.

- When in a offensive friendly environment, the team that is best able to capitalize on the easy runs is most likely to win. This benefits teams that are more tailored for run production in the stadium in play (HRs primarily, but other hit types that can be affected by the stadium effects apply, as well), usually the home team.

I then went back through the data I had to see if this was a noticeable trend and if it was something that was also explainable by overall record as well.

I made the assumption (based off the leagues I had data for) that in a typical OL approximately 70% of teams use a neutral to pitcher friendly stadium. So, following that, all else being equal, you're more likely to succeed with a neutral to pitching oriented ballpark as more teams draft that way, meaning, you'll win more road games and thus more games overall since they play more games against these opponents than the offensively oriented ballparks. However, even though it would seem that the opposite would be true as more teams draft offensive oriented ballparks, then also taking an offensive ballpark will improve your overall record as you'll win more road games, this would not be the case as offensive oriented ballparks as a whole differ greatly in how they boost your offense, so unless your ballpark is the same or similar (AFC, Wrigley, Tiger, etc) then even other offensive ballparks would be akin to visiting a neutral-pitching oriented ballpark.

So, in short, take the two bolded sections above and add the following for general rules of thumb on choosing a ballpark for your team (which I advocate choosing before building your team):

- A team that is built around an offensively oriented ballpark will do well at home, but will struggle on the road as their team will generally find it hard to score runs outside of the home environment they are tailored for.

- A team that is built around a neutral or pitching oriented ballpark will do well on the road, but will struggle at home as their team will generally find it hard to score runs in the runs suppressed environment of their home stadium.

Of course, there are exceptions to these rules in general, such as Petco oriented teams that can do well both at home on the road as their teams behave more like the offensive oriented teams at home as they have a unique ability to score runs in their home environment while suppressing their opponents ability to score while also taking advantage of the road advantage. Then, there's always teams that are just plain good all around or well-balanced that win everywhere, or the occasional exception to the rule.

Just to keep the important stuff altogether, here's the bolded stuff again all together with one more thought:

- When in a neutral or pitcher friendly environment, runs are harder to come by, so the team that scores first is more likely to win. This benefits the visiting team. A team that is built around a neutral or pitching oriented ballpark will do well on the road, but will struggle at home as their team will generally find it hard to score runs in the runs suppressed environment of their home stadium.

- When in a offensive friendly environment, the team that is best able to capitalize on the easy runs is most likely to win. This benefits teams that are more tailored for run production in the stadium in play (HRs primarily, but other hit types that can be affected by the stadium effects apply, as well), usually the home team. A team that is built around an offensively oriented ballpark will do well at home, but will struggle on the road as their team will generally find it hard to score runs outside of the home environment they are tailored for.

- More teams draft neutral-pitching friendly ballparks/teams, so it is currently advantageous to do the same. If/when the trend changes to more offensive ballparks, then tailoring your team to any neutral/offensive stadium and select pitching oriented ballparks will become the advantageous strategy.

This explains why I tend to do very well at home, but rarely win on the road, as I tend to select offensively oriented ballparks and teams, and why others see the opposite from their teams.
7/15/2012 12:56 PM
Insightful and fascinating. 
7/16/2012 8:21 AM
I've been saying this for a while, but for me it's simpler. But maybe I'm wrong because I usually come to a different conclusion. There is an inherent bias, and I'm not sure if it belongs to "us" (the collective players of WIS), or "Sparky" or some combination of both. 

We are inherently optimists. We set our bullpens, our lineups, our pinch hitters and our managerial settings the best way we know how. And when we do that, we are almost always setting them up for how we want things to work when we are winning.

So if you score the first run, then everything falls into place. Sparky uses your bullpen according to your optimistic roles. Your bench players get put in the position you expected them to be in when you win. 

Ergo... good hitting teams win better on the road because they are more likely to score the first run. Good pitching teams do better at home because they are more likely to hold off their opponent until they can score first. 

This is probably an oversimplification. Sparky isn't going to the bench or the 'pen after the first inning. Maybe the key is to be leading after 3 innings, or 5 innings. Whenever Sparky starts making decisions.
This post has a rating of , which is below the default threshold.
7/16/2012 1:59 PM
Posted by boogerlips on 7/16/2012 10:45:00 AM (view original):
If two teams are evenly matched the road team is only likely to score first because its the road team and they come to bat first. Any reasoning about being an "offensive team" or "defensive team" seems silly to me.

While it may be that sparky(you) manages slightly better with a lead, the odds of an evenly matched opponent being ahead of you after 7/8/9 innings is 17/14/12% in any park. It doesn't matter whether a lot of runs will be scored or only a few. All you have to do is divide 7/6, 8/7, 9/8. No need to overly complicate it. If you tailored your team to your ballpark better than your opponent that number goes down. Plus you obviously have the opportunity to score in the bottom of those innings. Plus its not like sparky only brings in good pitchers when you have a lead. I'd say your opponent should only win a game or two a year more than you at home, and thats before accounting for ballpark tailoring which should then give you a clear advantage.

A lot of people's problem with pitcher's parks is that they don't know how to draft for it. Mostly, I see people make mistakes on how many walks to buy. 
That was my initial thought all these years, that others just didn't know how to draft for the pitching parks. As for the rest, I may be entirely wrong, but it seems to hold up to general scrutiny as a premise based on an observation and an assumption with a couple of predictions as to what you'd see if it were true. Granted, I only had about 120 leagues worth of data to look at, which is far from enough to draw any real conclusions about home/road splits, but there were definitely trends and it was a larger sample than is generally offered in these discussions.

What I did want to add though was that the point wasn't so much about "offensive and defensive teams" as it was about the runs scoring environment. In a pitching environment where there is likely to be only 3.5 runs scored per game the team that scores first seems the most likely to win (and yes, the visiting team only has the advantage here because they bat first), but in a run scoring environment where there are 7 runs per game, the team that is most likely to capitalize on the easy runs is the most likely to win – and this is most likely to be a home team well-tailored.

Otherwise, as far as team construction and Sparky's managing, I would agree with you completely. I generally believe that most owners don't know how to draft for their home stadium and that most don't really know how to use their advanced settings to control Sparky's whims and otherwise odd decisions (usually supplied my mis-applied settings).
This post has a rating of , which is below the default threshold.
7/17/2012 3:20 PM
boogerlips, I'm fairly certain you're wrong about the odds.  Like 99.9999999999999...%
This post has a rating of , which is below the default threshold.
7/17/2012 10:12 PM
First of all, your "odds" of being ahead at the end of each inning are actually the relative amounts by which the road team should be ahead after batting.  And so the same percentages of more runs are more likely to be in units of full runs, IE a lead actually exists.  Also you completely ignore the fact that there is some standard deviation to the run scoring per inning, and that in higher-scoring environments this standard deviation is greater.  This tends to bring the probability of being behind down in higher-scoring environments.  Most importantly, your ridiculously over-simplified model assumes that the variable is continuously random and independent, and the whole point of the "Sparky isn't very good at managing from behind" argument is that if you are trailing when you come to bat in the bottom of the 6th inning and the opponent starts going to setup As and you start going to setup Bs and long men, then rather than having the simple regression towards 0 that you predict you have a higher probability of staying behind.  Even ignoring the inherent flaws in your numbers, that might mean that if you have a 17% chance of trailing after 7 it might look more like 15% than 12% in the 9th because if you're behind you're liable to stay behind.  By reducing it to a continuously random variable you don't actually disprove the theory at all, you just ignore it.
7/17/2012 10:15 PM
Posted by boogerlips on 7/17/2012 9:47:00 AM (view original):
I'm just not with you on the difference between 3.5 or 13.5 runs per game. I'm pretty sure the odds of scoring first and the odds of being ahead after certain late innings are exactly the same.

Thanks for taking the time running through 120 leagues of information and sharing!
If you thought about this for 30 seconds it should be clear that your logic is flawed if it predicts that the odds of scoring first are independent of run-scoring environment.  In, say, a 1.5 run/game environment there is just under a 60% chance of the home team scoring first.  In a 20-run/game environment it's over 70%.  Of course, those are based on the standard deviations in run scoring from RL, so it's pretty clear the numbers aren't perfect given that the standard deviations would almost certainly be much smaller in the former case and larger in the latter case.  But it shouldn't be off enough to bring them together in the middle.
7/18/2012 12:31 AM
Posted by dahsdebater on 7/17/2012 10:15:00 PM (view original):
Posted by boogerlips on 7/17/2012 9:47:00 AM (view original):
I'm just not with you on the difference between 3.5 or 13.5 runs per game. I'm pretty sure the odds of scoring first and the odds of being ahead after certain late innings are exactly the same.

Thanks for taking the time running through 120 leagues of information and sharing!
If you thought about this for 30 seconds it should be clear that your logic is flawed if it predicts that the odds of scoring first are independent of run-scoring environment.  In, say, a 1.5 run/game environment there is just under a 60% chance of the home team scoring first.  In a 20-run/game environment it's over 70%.  Of course, those are based on the standard deviations in run scoring from RL, so it's pretty clear the numbers aren't perfect given that the standard deviations would almost certainly be much smaller in the former case and larger in the latter case.  But it shouldn't be off enough to bring them together in the middle.
I'm confused; in a 20-run/game environment, the HOME team has a 70+% chance of scoring first?  It seems it would be the other way around; with each team averaging over a run a game, the team batting first would have the greater chance of scoring first.  I'm no mathematician, though ...
7/18/2012 2:31 AM
Sorry, you're right of course...
This post has a rating of , which is below the default threshold.
7/18/2012 9:04 PM
Those aren't odds of being ahead, they're theoretical ratios of runs.  If they're ratios of larger numbers of runs then it's more likely to translate into real leads.
This post has a rating of , which is below the default threshold.
of 2
All Forums > SimLeague Baseball > SimLeague Baseball > Home Field Advantage

Terms of Use Customer Support Privacy Statement

Popular on WhatIfSports site: Baseball Simulation | College Basketball Game | College Football Game | Online Baseball Game | Hockey Simulation | NFL Picks | College Football Picks | Sports Games

© 1999-2014 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.