We figured out that playing against a bad defense can give a 15% advantage, and against a good defense a 15% disadvantage. But how does that work for home vs away, the day of the week, and other factors?
First let's look at the data: I took again the QB data for the top 10 QBs who played 16 games, to have a good comparison. We had already found that playing against a defense makes a difference; once we filter that effect out, can we find other effects? I looked specifically for:
- home vs away
- day of the week (Sunday vs Monday vs Thursday)
- days of rest between games
QBs on average perform a bit better at home than away. The difference is not as large as for the defense factor, but it is there.
For Monday it shows what we expect; for Saturdays I had only 2 games, so I shaded that bar as I think we need to ignore that, but Thursday actually shows the largest positive effect. That's surprising to me, so let's look at the games.
It seems that the Thursday games my subjects were involved in, weren't that bad at all. I marked the games where they underperformed as red, the "normal" performances as blue, and the overperformance as green.
There's lots of green - yes, there were some crap games (eg the game in week 1), but overall it wasn't too bad. Even when you correct for home/away, there is still an advantage to be seen.
So even though I'm still not 100% convinced, it seems from my limitede data that both Thursday and Monday games have a sizeable advantage. More advantage than home/away, but less than good/bad defense.
This is unexpected. QBs on a short week rest perform better; on 2 weeks they perform wors; but on a Monday or Sunday-Thursday (10 days) break again they perform better.
This seems to be mostly an effect of the Thursday/Monday games, although it's interesting that on a long break (14 days) the QBs perform worse. A potential explanation is that after 2 weeks the defense is well rested, and they would benefit more from rest than the QBs who have a less physically demanding job.
First let's look at the data: I took again the QB data for the top 10 QBs who played 16 games, to have a good comparison. We had already found that playing against a defense makes a difference; once we filter that effect out, can we find other effects? I looked specifically for:
- home vs away
- day of the week (Sunday vs Monday vs Thursday)
- days of rest between games
Home vs Away
This is a relatively clear; and the data confirms what we know.
QBs on average perform a bit better at home than away. The difference is not as large as for the defense factor, but it is there.
Day of the Week
We all know that players give a bit extra on Monday night, and Turdsday games are crap. Does the data show that?
For Monday it shows what we expect; for Saturdays I had only 2 games, so I shaded that bar as I think we need to ignore that, but Thursday actually shows the largest positive effect. That's surprising to me, so let's look at the games.
It seems that the Thursday games my subjects were involved in, weren't that bad at all. I marked the games where they underperformed as red, the "normal" performances as blue, and the overperformance as green.
There's lots of green - yes, there were some crap games (eg the game in week 1), but overall it wasn't too bad. Even when you correct for home/away, there is still an advantage to be seen.
So even though I'm still not 100% convinced, it seems from my limitede data that both Thursday and Monday games have a sizeable advantage. More advantage than home/away, but less than good/bad defense.
Rest days
One of my theories is that players will play better on more rest days. However, this is not exactly what the data shows:
This is unexpected. QBs on a short week rest perform better; on 2 weeks they perform wors; but on a Monday or Sunday-Thursday (10 days) break again they perform better.
This seems to be mostly an effect of the Thursday/Monday games, although it's interesting that on a long break (14 days) the QBs perform worse. A potential explanation is that after 2 weeks the defense is well rested, and they would benefit more from rest than the QBs who have a less physically demanding job.
Conclusion
- Home vs away is a small difference of less than 5%
- Monday and Thursday is an advantage of over 5%
- QB's on a 2 week rest may score a bit less. The same goes for the first week of the season.
Comments
Post a Comment