Home field advantage

[Curtis]

Originally Posted by statfreak

Once the game enters the bottom of the last inning everything else (other than the score) gets thrown out the window. The home team now decides the final game result. The advantage comes when the home team scores more runs than the visitor and thus immediately ends the game. Yes, both teams may have had an equal amount of outs to score runs; however, only the home team has the ability to end the game.

And how is that relevant to the topic of home field advantage? The visiting team has the advantage of more outs for the first 8 1/2 innings, the home team has an advantage from the bottom of the 9th on. Logic says it should even out.

Originally Posted by Henry

The ONLY time in the entire game that the specific amount of effort to win the game is known is in the bottom of the 9th or following innings - and the Home team is the only one that has its offense in play during that time. As a visiting team, you never know if what you just did (scored 1 or 20 runs) is enough - but as the Home team you know exactly what will end the game immediately.

In football, they not only trade the sudden death advantage, but they know what it takes to win the game and are in control of it at the time. In baseball, only the Home team has that option and only in their half of the final inning.

Sudden death isn't relevant to this (baseball) situation. It WOULD be relevant if the game ended as soon as either team scored. As it is, each team is treated equally in extra innings, as in college or high school football.

Originally Posted by fintach

I mentioned this above, but it might have gotten lost in the ::cough::excessive::cough:: length of my previous post.There may also be a psychological difference of approach, even in the managers. The old baseball adage is "Play to tie on the road; play to win at home." Considering the number of managers who buy in, at least a little, to conventional baseball wisdom, that may affect a manager's approach.

You may be right about the psychology.

I have never understood the adage you quote, and for pretty much the reason you're assuming. Why would I, as the visiting manager, play to tie, knowing that in doing so I would be giving you the chance to score one run unanswered. If I play (successfully) to win, then you as the home team need to put up 'a crooked number', which is MUCH less likely.

Originally Posted by fhomess

It would seem to me that knowing how many runs you need to win as the home team in the bottom of the 9th isn’t a particularly big advantage. The batting team has an inherent disadvantage in that the batting order is locked. They cannot simply bat their three best batters. They have to bat whoever comes up or a pinch hitter. Generally speaking, your best hitters are already in the lineup. In a tight game, it’s also likely that you’ve already used your better pinch hitters in the 7th or 8th inning. The defensive team, on the other hand, has the ability to bring in relievers and adjust their defense if they have good fielders available. Given that the visiting team has had 3 more outs than the home team by the time the bottom of the 9th rolls around, and the fact that the home team likely only uses their best relievers in the event of a lead, I would expect that they’re more likely to get there in the lead than tied. So the visiting team has the advantage because they have had more chances and a more predictable reliever usage pattern.

By George, I think you've got it!

Originally Posted by Kobeck

I am currently reading Baseball Between the Numbers. This issue is discussed, indirectly, in a number of ways. Like all things in baseball there does seem to be a 'home advantage' - as descriped earlier in the thread, but it is not as big as is widely believed.

If the bottom third of the order was up and only 1 run was needed too win, those defensive specialists would probably be PH for, or not if one of them got on second with no outs then the one or two situations where the sacrafice is beneficial would be available. If that runner got to third with 1 out the visitors could IBB the bases full and play for the force at home, but then the fielders would be in a few steps and what otherwise would have been an out is tranformed into the game winning single ect...

Certainly without doubt the home team has an advantage in the bottom of the 9th and onward, but it is not a huge advantage.

… and ditto.

[Henry]

Originally Posted by Curtis

Sudden death isn't relevant to this (baseball) situation. It WOULD be relevant if the game ended as soon as either team scored. As it is, each team is treated equally in extra innings, as in college or high school football.

I think you missed the point I was trying to make by defining Sudden Death based on what it is in football. The point I was making is that while sudden death is available to BOTH teams in football (=to whoever has the ball), it is ONLY provided to the Home team in baseball - and ONLY from the ninth inning on.

A lead built by the visiting team at ANY point in the game does not automatically win it without the Home team having a last chance to answer. Any lead built by the Home team however is a game winner from the 9th inning on.

Small advantage? Yes, but enough to win at a historical average of .540

One of the biggest assumptions made about baseball that results in bad analysis is that everything is fair since both teams get 9 innings. Although it's true they both get 9 innings, it ignores the fact that if a team scores 700 runs over the season, they are not accumulated at 0.48 runs per inning - they are scored in various amounts in between many innings in which no runs are scored. This small detail means if runs could be traded between games to obtain more wins (ie: split up a 7 run inning across 3 one run losses to gain 3 wins) we see that singular runs can make a huge difference in win-loss records. Being the home team in the bottom of the ninth or in extra innings of close games is a much better place to be than gaining the go-ahead run as the visiting team and having to get 3 more outs.

You may not agree, but as I said earlier - I much prefer last-at-bat as the primary reason for home field advantage than some loosely held belief about home field familiarity. And besides, a 4% difference across 162 games just isn't that much.

[Curtis]

Henry, what you don't seem to be seeing is that the visiting team has the advantage up until the bottom of the ninth. Why do you think the home team's advantage after 8.5 outweighs the visiting team's advantage before then?

You're trying to turn this into a matter of opinion when it's a matter of fact; we aren't just blue-skying notions here. This has been studied, your hypothesis was tested, and it was found to be unsupported.

That your hypothesis is not logical is just icing on the cake.

[Le Grande Orange]

Originally Posted by Henry

Small advantage? Yes, but enough to win at a historical average of .540

You have not provided proof for this argument. Again, how many games have been won in either the bottom of the 9th inning or extra inning?

[Curtis]

Originally Posted by Le Grande Orange

You have not provided proof for this argument.Again, how many games have been won in either the bottom of the 9th inning or extra inning?

And how many by the visiting team?

[Henry]

Originally Posted by Curtis

Henry, what you don't seem to be seeing is that the visiting team has the advantage up until the bottom of the ninth. Why do you think the home team's advantage after 8.5 outweighs the visiting team's advantage before then?

Maybe because Home teams win 54% of the time? Whether I'm right or wrong there has to be a reason, and I'm just saying no one has "proved" what constitutes that advantage. There have been plenty of studies stating certain ideas were not validated, but none that have uncovered an alternative option as the reason. As long as that proof is not available, it's reasonable to assume all the analysis to this point in time missed something and all theories are still possible.

Originally Posted by Curtis

You're trying to turn this into a matter of opinion when it's a matter of fact; we aren't just blue-skying notions here. This has been studied, your hypothesis was tested, and it was found to be unsupported.

All we have is opinion at this point - otherwise some study would already be out saying "here's the reason for the 0.40 home field advantage". There's lots of theories but no final anwser.

Originally Posted by Curtis

That your hypothesis is not logical is just icing on the cake.

I'm sorry you feel that way. I tried to explain my logic, but your obviously not seeing it. maybe this will help - think of it as a bowling game. You and I bowl 10 frames and you finish with a 205. If I get 6 pins on my last ball I finish with 206 and win. You don't get another chance or frame and we both had the same opportunities - the advanatge I have is knowing exactly what I need to win or lose - when you rolled your 10th Frame you were going for the best you could get but had no idea whether it would be enough or not - I do.

Originally Posted by Le Grande Orange

You have not provided proof for this argument. Again, how many games have been won in either the bottom of the 9th inning or extra inning?

Again, I'm not trying to offer proof - only why I believe in a specific theory. Given the fact that one would have to summarize somewhere around 250,000 games since 1900 to get that number, few of us have the time to attempt it.

Here's a question... did any of the studies that claim this theory has no merit do the analysis your asking for? If yes, my theory is void. If not, maybe the theory is still alive.


By the way guys, no one has to be right here - I thought this was a discussion board?

[Henry]

Ok, spent half the night looking for other work to see if anything would help the discussion. Surprisingly, I found something that on the surface migh play against my theory...

Matt Swartz compiled a chart for runs scored each inning, the percentages are the difference in runs scored for the home team vs. the visiting team:

1 16.2%
2 9.3%
3 10.1%
4 6.0%
5 7.8%
6 8.1%
7 8.7%
8 6.5%

Per his calculations, the average per game is about 8% more runs scored by the home team - that supports the .540 historical winning percentage. Although this doesn't prove WHAT causes the anomaly, it does suggest that most of this difference occurs in the first three innings (although the 9th inning can't be calculated since the visitor-home opportunity isn't equal due to the Home team not batting in the 9th in games won at that point).

At this point, I thought my theory was completely out the window - but, like others, the next question is why the first three innings? What can possibly be different in those innings than the others? Familiarity doesn't fit, otherwise it should be spread out across the game more.

I then began to wonder (and waited till I got some sleep to see if it still made sense lol) if the fact that a home team knows/believes it has an advantage, set out to get the lead early thinking THAT plus batting last tilts the game in their favor? If so, HFA is truly a psychological advantage which in real life statistics just causes the team to try 110% harder (actually 108% ) in the early innings.

I'm not sold on this yet - and the fact that the 9th inning can't be calculated here doesn't help me get there. Swartz, himself, offers no conclusion - simply presents the numbers and says further study is needed - but I'm still going to have a hard time accepting that going into the bottom of the 9th or later, knowing I own a sudden death scenario, isn't going to also raise my efforts to their maximum. It does appear, however, I'm going to have a harder time with that considering the .540 advantage may mostly exist in the early innings of a game.

Maybe it's just the fact that the home team "believes" last-at-bat is an advantage that causes the extra effort - - - or maybe the extra effort is an unconscious attempt to avoid batting in the bottom of the ninth.

Ok, I'm tired...

[The Wolf]

Could the first three inning effect be a function of simple acclimatization? The home team is very used to playing on their own field, and the visitors aren't, and it takes the visitors a few innings to get familiar and comfortable with the lighting, geometry, pitcher's mound, etc.?

[fhomess]

Originally Posted by The Wolf

Could the first three inning effect be a function of simple acclimatization? The home team is very used to playing on their own field, and the visitors aren't, and it takes the visitors a few innings to get familiar and comfortable with the lighting, geometry, pitcher's mound, etc.?

I remember seeing an article at some point (I was unable to find it just now with a quick googling) where the author showed statistically that home teams won the first game of series at a higher rate than the 2nd or 3rd game in the series. This leads some credence that the travel and familiarity does indeed have an effect on winning games.

The other thing that I find fascinating about baseball is that despite the non-uniformity of fields and the home team batting last, baseball teams have a smaller home field advantage than the other major sports (NBA is up close to 60%).

[fhomess]

Ok, since I'm the type of person who's happy to give a quick pass at some statistics, I went to Retrosheet and downloaded the game logs for the past 10 seasons (2001-2010). Below is a table of win percentage based on home or away status for each season, as well as a totals line. The purpose of this very brief study isn’t to determine where home field advantage comes from, but to determine whether or not batting last is potentially a contributor to home field advantage.

For each season, I tallied up the wins by the home and away teams to determine the home field advantage in all games. Then I did the same thing only for extra inning games. The data doesn’t make it particularly easy to determine games that go into the 9th inning tied, so I just used extra inning games. Any batting last advantage that appears in the 9th inning ought to also appear in extra innings, so I don’t feel badly about excluding the 9th.

The data below tells us that the home field advantage over the past 10 seasons has led to a .550 winning percentage at home. This is slightly higher than the accepted historical value of .540 that people have been bandying about. The second pair of columns shows the home field advantage in extra inning games. Initially, I was expecting that extra inning games would be won more often by the home team if there was a true advantage to batting last, even more so than the advantage over all games. At minimum, I was expecting it to be identical to the home field advantage over all games. This was not the case over the past 10 seasons. The away team won more extra inning games over these years than expected, and this was true in every individual season except 2006 and 2002. Yes, some of those seasons are close, but some aren’t. Each season typically saw about 200 extra inning games, or about 8.5% of all games, which is a total of 1028 extra inning games in the study.

One thing I thought was fun is that the 2007 San Diego-Colorado game 163, 2008 Minnesota-Chicago game 163, and 2009 Detroit-Minnesota game 163 were all won by the home team, with two of those games going to extra innings. All 3 are included in the study.

Year  AWins  HWins    HFA   xAWins   xHWins   xHFA    Diff
2010   1072   1358   .559      104      116   .527   -.032
2009   1097   1333   .549       89      106   .544   -.005
2008   1077   1351   .556      100      108   .519   -.037
2007   1113   1318   .542      103      117   .532   -.010
2006   1102   1327   .546       80      105   .568    .021
2005   1124   1306   .537       85       97   .533   -.004
2004   1129   1299   .535      103      115   .528   -.007
2003   1094   1335   .550       89      108   .548   -.001
2002   1111   1314   .542       85      115   .575    .033
2001   1155   1273   .524      101       94   .482   -.042

Tot    5461   6687   .550      476      552   .537   -.013

x-Columns denote extra inning games

[fhomess]

Here's the same table extended to 1991 (20 seasons)

Year  AWins  HWins    HFA   xAWins   xHWins   xHFA    Diff
2010   1072   1358   .559      104      116   .527   -.032
2009   1097   1333   .549       89      106   .544   -.005
2008   1077   1351   .556      100      108   .519   -.037
2007   1113   1318   .542      103      117   .532   -.010
2006   1102   1327   .546       80      105   .568    .021
2005   1124   1306   .537       85       97   .533   -.004
2004   1129   1299   .535      103      115   .528   -.007
2003   1094   1335   .550       89      108   .548   -.001
2002   1111   1314   .542       85      115   .575    .033
2001   1155   1273   .524      101       94   .482   -.042
2000   1116   1312   .540      101      101   .500   -.040
1999   1163   1264   .521       98       93   .487   -.034
1998   1123   1307   .538      105      102   .493   -.045
1997   1053   1213   .535       90      106   .541    .006
1996   1039   1227   .541      110      108   .495   -.046
1995    944   1072   .532       81       99   .550    .018
1994    773    826   .517       68       75   .524    .008
1993   1047   1221   .538       97       99   .505   -.033
1992    944   1162   .552      109      101   .481   -.071
1991    973   1131   .538      103      117   .532   -.006

Tot   21249  24949   .540     1901     2082   .523   -.017

[Henry]

fhomess,
...make sure I'm understanding the table right...


xAWins is extra innings wins by the Away team and xHwins is extra inning wins by the Home team? If so, the 2001-1010 table shows a .537 extra inning HFA compared to .550 overall, less than expected (which I think is what you were saying).

In the 1991-2010 table, its .523 compared to an overall of .540 = same result effectively.

Your comment about the Away team winning more games, however, has me confused - in the 20 year table, Home team won 2,082 xinnings games compared to 1,901 for the Away team. that's 181 more games for the Home team over that period - not significant, but still a home team advantage.

[fhomess]

Originally Posted by Henry

fhomess,
...make sure I'm understanding the table right...


xAWins is extra innings wins by the Away team and xHwins is extra inning wins by the Home team? If so, the 2001-1010 table shows a .537 extra inning HFA compared to .550 overall, less than expected (which I think is what you were saying).

In the 1991-2010 table, its .523 compared to an overall of .540 = same result effectively.

Your comment about the Away team winning more games, however, has me confused - in the 20 year table, Home team won 2,082 xinnings games compared to 1,901 for the Away team. that's 181 more games for the Home team over that period - not significant, but still a home team advantage.

Your interpretation of the data is correct. My comment should be read as "the away team won more extra inning games than expected".

[Curtis]

Thank you for your research, fhomess.

[Le Grande Orange]

Originally Posted by Henry

Maybe because Home teams win 54% of the time?

You're overlooking something: that figure is the overall average for the entire league, combining the records of all clubs. If you look at individual clubs, it's a different story. In 2010, for example, Atlanta was .691 at home; Minnesota, .654; Philadelphia, New York Yankees, St. Louis, Colorado, and Detroit, all .642 at home. Are you trying to suggest that all those clubs achieved such a high winning percentage at home due to winning games in the bottom of the 9th inning or extra inning? (Incidentally, in the 2010 season 22 of the 30 MLB clubs posted a .500 or better record at home. 28 clubs had a better home record that away record; 1 club had the same record at home as it did on the road; and 1 club won more games on the road than it did at home.)

Originally Posted by Henry

Again, I'm not trying to offer proof - only why I believe in a specific theory.

But your theory, if I understand it correctly, can either be supported or refuted by looking at the results of major league games. So why not at least a cursory examination of the available data to see if your theory fits?

[Henry]

LGO, I believe what fhomess posted above is what you were asking for earlier - and from what I can surmise it doesn't support what I proposed. His 20 year table says that home teams overall from 1991-2010 won 54% of their games, but only 52.3% of their extra inning games (minus the 9th inning data). Add that to what I posted earlier that the largest HFA numbers occur in the first three innings, and my theory does look pretty weak.

As much as I hate to admit it, it appears the HFA is either a home field familiarity issue - or a psychological one based on the common belief that home teams have the advantage, thus causing it to be true because of an added unconscious (and undocumented) effort.

In summary, I have to admit I am likely incorrect. In any case, I think I would rather be the home team anyway

[Vinny P.]

There is a psychological effect of "familiar surroundings," of playing where you practice.

Many interesting comparison studies have been done with students taking a test in the same classroom in which they learn, and students taking the same test in a different room altogether. Guess which group of students consistently score higher?

[Kobeck]

First the talent of a well ran team should be customized for the home field giving the team a built in HFA and any visiting team would have to scramble to match those skill sets.

For instance a team with a huge LF would essentially need 2 CF types to patrol the OF, whereas a team like the Red Sox would be at a disadvantage going into that park. Of course there are numerous variables involved but any front office with any talent should get guys that could outperform in the park they play 81 games in.

Second the way I see the data it confirms the slight advantage the home team has from bottom of the 9th on - IMO. Those extra inning games are played by evenly matched teams or else they would no go that far all tied up. The home team consistently winning more than half of those 'sudden death' games shows that there is at least a slight advantage, at least too me.

[fhomess]

Just to be clear, the data does not say that home field advantage disappears in extra innings. It does suggest that home field advantage is slightly lower in extra innings, and from that I conclude that batting last is not a factor in home field advantage. I don't make any other conclusions from my data about what contributes to home field advantage. I think people have hit on a number of those points: team construction, familiarity with the surrounds, potential umpire bias, psychological effects, strategic effects that may not be warranted based on who bats last, etc. It's probably a combination of all those things, and I do think that there's a HFA no matter how long the game lasts.

[Le Grande Orange]

Originally Posted by Kobeck

First the talent of a well ran team should be customized for the home field giving the team a built in HFA and any visiting team would have to scramble to match those skill sets.

That may explain a portion of it. But note that home field advantage exists in the minor leagues as well, and it is far less likely a club there is tailoring its roster to its park.