THE BOOK--Playing The Percentages In Baseball

KC Joyner claimed he did this for the NFL.  Charting every play was the basis for his Scientific Football books.

I’m skeptical that you could watch a game from any sport in 30 minutes unless you were somehow able to automate the fast-forwarding.

This type of thing is what I hope metrics moves towards.

I recall seeing Curt Schilling had DVDs of everpy PA versus all of the players on the upcoming oppent on DVD.

I brought this situation up at FG in regards to whether Abreau is getting lucky/unlucky on BABIP.

If we could watch all of his AB’s, the answer would likely be conclusive and we could stop just conceding something up as “luck” or “unluck” just because we don’t have the resources, time, or desire to find out.

My guuess is that sometime in the future, we’ll be able to sort end of game highlights by player, and then be able to go and look at a single player’s ABs for a month ata time, or the entire season, and be able to actually see what we are trying to measure.

IMO, we’re to the point to where resources limit what is being done with metrics. It is not likely that the same metrics can be viewed in a different and novel way to reveal something new and worthwhile. What we need is more data, observations, etc.

What is amazing to me is just how fast you can watch any game, when the in between play stuff is removed, commercials, etc. It is literally amazing how “little” action there is in watching an NFL game.

@circlechange

the bloomberg sports demo at the fangraphs live event demonstrated this kind of capability.  you could sort through all of the video for every single pitch a batter saw based on the pitcher, count, result, pitch location, pitch type and a few other things.

i think it is something that they are only marketing to pro teams right now though.

@Jim A

As someone who has charted football games in the same manner as KC Joyner, I can tell you it takes between 3-4 hours to chart one game. As I have also pitch charted baseball games professionally, it takes about the same amount of time for one game of that as well.

You can certainly watch a full game in less than an hour, but you can’t record data during that short span of time.

I have talked to KC Joyner about how he charts football games and he doesn’t fully chart every play of every game each week. That is an impossible task for one man. So he picks through games and records data that he needs for that week and then finishes games in the off-season.

Jay/4, that was exactly my thought.  When I’ve charted pitches, it takes me much longer to record the data than it did to watch the pitch happen.  And often I have to rewind and watch it again to get something I missed the first time around.  It ends up taking on the same order as the duration of the game in real time, if not longer, depending on what I’m charting.

If you just wanted to mark batted ball locations or something like that, you could cut out a lot of plays and do it quite a bit quicker, I suppose.

Tango addressed his arithmetic to Cy’s question, which of course had nothing to do with charting pitches or plays. I can’t imagine watching that many games and getting anything factual out of it without recording something as you were going.

When I’m watching a game for entertainment, I’m happy with the pace as it stands, except for between innings.  When I’m trying to watch a game as fast as I can, it’s usually because I’m charting something.

You’re right, that you could not record much data or chart much while watching that fast. Maybe if the games came prepackaged so all you see is each pitch, you might be able to record your thoughts and observations into an audio recorder.

It could be too much information and maybe no one, not even a seasoned writer or a former manager would be able to make sense out of it. I guess if this is the case, it just tells us how important data is.

I also wonder how this is related to MVP voting. Maybe some AL teams only play the Rangers 6 times. How can a writer from one of those cities make an informed vote on Josh Hamilton, for example. Maybe they see him a few times on TV. I think they have to turn to data.

Don’t forget that I was only suggesting one league, not all of MLB. Maybe an hour per game could be enough to think about/record some observations about what you see. That would mean 7 hours for the AL and 8 for the NL.

Right, my post was purely for watching it. 

From a viewing experience, any dead time is lost time.  So, between inning time, batter getting in and out of the box, pitcher taking his time throwing pitch, pickoffs without consequence. 

If we remove dead time, the game would last about 2 hours, which is the ideal time for me. 

Same for hockey, that if you remove dead time, you are down to 2 hours.  This is proven with Olympic hockey, where the pace is relentess, the breaks are very short, and the game is over in 2 hours.  It’s beautiful.

As for charting data, I’m sure you’d need at least 3 hours to record a game.

As David Biderman from the WSJ calculated it, (http://bit.ly/8hCzWL) NFL games only have 11 minutes of actual gameplay. With 32 teams playing 16 games each, that’s about 47 hours per season, or a little under 3 hours per week.

You could learn quite a lot! Although at the 11 minute pace you’d have to have a fast mind to take anything in.

@dbc

Yes, at this point int itme, the only personel to be able to do would be a stats guy employed by a team. I doubt anyone else would have the funding or time (or desire) to put in all that work.

I believe that Will Carroll watched every pitch in an MLB season a few years back.

This is certainly the type of task that requires a passionate individual, for to ASSIGN it as a task to an employee is to invite a lawsuit. It brings to mind the guy who 1) arranged to videotape every channel on his cable system for one day, 2) spent the next several months watching all of that tape, 3) spent some time following the watching by climbing some mountain, and 4) wrote a book about the experience (can’t recall his name—sorry). If some good reading results from the experience, then I’d encourage an interested party to go for it.

#10: “Same for hockey, that if you remove dead time, you are down to 2 hours.  This is proven with Olympic hockey, where the pace is relentess, the breaks are very short, and the game is over in 2 hours.  It’s beautiful.”

It sure is, Herr Tango. But you can’t sell as much beer and other concessions, and thus the beauty gets purposefully uglified. Sigh.