Home advantage and capping.

[Lori]

First off, anyone who has been sports betting for any amount of time is likely to learn nothing from this post. I'd appreciate it if they read it though, maybe they could teach me a thing or two.

Something I've been fighting for some time now is the instinct to work out home advantage and then factor it into my capping.
It's fine, and likely neccessary, if you're using a mechanical system, however when you're using a mixture of stats and observing what is going on, it's actually a strange concept when you boil it down.

What is slightly perverse is the following.... I suspect at least 90% of the people reading this have done this before (I certainly have.)
It's typical to break down the two teams, see who you think is the better, make a line and then add a number according to home advantage.
This seems all very logical of course, until you think about what you're doing.
In the process of working out which team is better, you've had to eliminate home advantage from all your working (again, this is understandable in a purely mechanical system), work out who is the better team on a neutral field (which accounts for 2-3 games a season in football) and then add it back on. I'm going to hypothesize that humans are bad at this. (As can be seen by typical BSP scenarios)

Something I've found that has aided the non-mechanical side of my capping is to think about how teams play at home and the road, without assigning a specific number to it. If the Chiefs are playing at home to Denver, I consider games where I've seen the Chiefs at home, and Denver on the road. Obviously the conclusion I come to will be 6-7 points different (or possibly 8 or 9 in that exact extreme example) to the one I would come to if the roles were reversed, but the difference is that I'm not making assumptions twice... assumptions that at best cancel out.

Of course, I'm still using info gained from the games at the opposite fields to the game I'm capping, but I'm using those as a guide and to increase my sample size. I'm mentally attributing less weight to those games.

The reason I've started doing this more and more this season is quite simple. Not all points have the same value.
You could be in the position where a team is a 4 point dog on the road and a 2 point fave at home. Have you really done that team justice?

Similarly you could turn a 7.5 point road fave into a 13.5 point home one. All very nice, but have they gained anything?

On the other side of the swing, you could move a 8 point road dog into a 2 point home dog.

The above three examples are all viable, however it should be clear that they don't all carry the same strength.

I've found I've understood the game better since I considered teams at home and away as similar, but seperate entities where one helps you to increase your sample size and modify the other, rather than removing and adding points in a somewhat arbitary fashion.

All thoughts appreciated.

Lori

[rjp]

It's good that you realize home field advantage isn't something you can express as points (especially in football), as the numbers just don't add up right. In some cases you're giving too little and in others too much. Taking this a step further and treating all games as a winning percentage and you should be able to quantify home field advantage much better than the average Joe.

That said, home field advantage isn't something I've studied all that deeply. What I do know, however, is that it might bet better to look at it as away team disadvantage. Traveling and staying in hotels in an unfamiliar place takes its toll. It isn't always about how loud the home crowd cheers (or boos)....

[MicroBob]

"It isn't always about how loud the home crowd cheers (or boos).... "


QFT.
As I mentioned in another thread, I did my fair share of traveling around in mid-major womens basketball. Crowds of 300 or so in 8k+ seat arenas.
But I would estimate that home-court advantage was still worth roughly 3 points even in that kind of league. This was especially easy to see because they would play most of their in-conference opponents at home and on the road and you could just look at the results.

Not necessarily comparable to the NFL of course. But it's all a part of human and athlete psychology I believe.

[Rustjive]

[ QUOTE ]
What I do know, however, is that it might bet better to look at it as away team disadvantage. Traveling and staying in hotels in an unfamiliar place takes its toll. It isn't always about how loud the home crowd cheers (or boos)....

[/ QUOTE ]

How meaningful is that (not rhetorically)? For many cases, there may indeed be no home field advantage, although it can't really be an overwhelming majority. Home Field Advantage at places like Arrowhead and Lambeau are well documented, and while it may not be particular to a certain franchise (ala x plays well at home because of fans), teams are sometimes built for the type of physical environment (dome/turf combination) the venue offers.

The thing I really mean, however, is that the NFL in the vast majority of scenarios have a binary input - you are either playing at home or you are playing at the road. If we treat it as a road disadvantage instead of a home field advantage - po-tay-toes, po-tah-toes, right? Then add the possibility of there actually being real home field advantages as mentioned above, then what did we gain moving off the 3 point baseline? I certainly don't deny that HFA is complex, has tons of factors and generally the 3 points doesn't do it justice, but you have to work from somewhere, no?

[rjp]

Rustjive, I just wanted to point out that there's much more to home field advantage than the infamous "12th man".

Teams built for specific environments is a very good point and should be taken into consideration. This obviously doesn't hold for basketball, but a good thing to know for football.

[Thremp]

[ QUOTE ]
It's good that you realize home field advantage isn't something you can express as points (especially in football), as the numbers just don't add up right. In some cases you're giving too little and in others too much. Taking this a step further and treating all games as a winning percentage and you should be able to quantify home field advantage much better than the average Joe.

That said, home field advantage isn't something I've studied all that deeply. What I do know, however, is that it might bet better to look at it as away team disadvantage. Traveling and staying in hotels in an unfamiliar place takes its toll. It isn't always about how loud the home crowd cheers (or boos)....

[/ QUOTE ]

A flat % adjustment is errant as well (Not that you suggest this method) since HFA is less important for teams that are huge faves or dogs already. You might actually be better off in a sport (speculation from here out) doing a cents adjustment to a line for something like baseball rather than a % adjustment.

[digdeep]

Great post Lori... I think one thing to consider is the variance of a team's play. Those teams with higher variance in their play deserve a bigger edge at home vs. on the road. I.e. PIT this year in the NFL - a completely different team at home vs. on the road.

The teams that have low variance, which are generally the best teams and worst teams likely deserve less of a home/away adjustment as they are far more consistently good or bad.

Would peope generally agree/disagree?

[Runner Runner]

I disagree digdeep. I actually think the opposite is true more often then not.

Variance in play or inconsistent play will bring a team closer to mediocrity. It will make a good team worse and a bad team better. It will bring them closer to .500 and more likely to cover as a dog or not cover as a favorite (unless this is already factored into the line). I think that for teams that are just flat out inconsistent for any reason, home field advantage/disadvantage will actually apply less, and what they bring to the table that day will more then likely be the determining factor.

[1MoreFish4U]

Good & interesting thread.

[digdeep]

[ QUOTE ]
I disagree digdeep. I actually think the opposite is true more often then not.

Variance in play or inconsistent play will bring a team closer to mediocrity. It will make a good team worse and a bad team better. It will bring them closer to .500 and more likely to cover as a dog or not cover as a favorite (unless this is already factored into the line). I think that for teams that are just flat out inconsistent for any reason, home field advantage/disadvantage will actually apply less, and what they bring to the table that day will more then likely be the determining factor.

[/ QUOTE ]

Your point makes logical sense. I was hoping to find a relationship where there likely isn't...

I quickly ran a correlation analysis, and your are right that there didn't seem to be a strong correlation with variance and home / away winning %.