You'll have to forgive this as it has little to do with the Nationals. Well, nothing actually, except they play baseball and this is about baseball. I could write another post professing my never-ending Man Love for Livo but I think my small cadre of followers gets the message there. I could write the Pirates another thank you note for Mr. Capps but I don't want to get too cocky in April.
Instead, I want to send a bill to the Sabermetrics folks for the truckload of Advil I just ordered. MY HEAD HURTS. STOP IT ALREADY.
I've poked fun at what I call Slide Rule Baseball before and most of it is just that. The fact that I don't understand all these convoluted, made-up stats doesn't mean they aren't legitimate. It means you have to be a lot smarter than I am to figure it all out.
But this took me over the edge.
Some background - I spend time at a site most for sports journalists called, yeah, SportsJournalists.com. I'm one of the moderators. The site has been a great resource for me. I've made some friends from the site, seen my network grow considerably. And lost a lot of hair.
Lots of SABR types there, too, and some were out in force when the discussion turned to Ryan Howard's new deal. For the record, I think it is excessive because I don't think anyone is worth that kind of money. I also had no idea until my first game this season that Howard was 30. For some reason, I thought he was 26 or so.
I do concede he's one of the better players in the game, something some of the SABRs are arguing otherwise. Though one of the arguments claims convoluted stats show he's 40th best. Really? You know, that pretty much puts him in the top five percent. I think if you are better than 94 out of every 100, you are one of the better players.
I cut this verbatim from one of the posts. The guy who posted it is a good guy, just a little over the top with his slide rule stuff. The following is why I need Advil by the score:
Wins Above Replacement is derived by taking the average number of runs per win in baseball and dividing that into Runs Above Replacement. RAR is calculated by taking the following: Batting Runs Above Average, Fielding Runs Above Average, a positional adjustment and a replacement level adjustment.
Batting Runs is derived from a linear weights formula (basically, they do a regression analysis to decide how much, on average, a batting event is worth. On average, a HR is worth 1.7 runs more than an out, a single is worth 0.77 runs more than an out, etc. Then a player is credited each time he does those things with the determined run value).
Fielding Runs is derived from a stat called Ultimate Zone Rating, which credits fielders with making plays on balls hit in their area and debits them for plays tha't don't get made.
A positional adjustment is the average number of runs the average player at that player's position is worth compared to the average hitter in general. A first baseman or corner outfielder will get dinged, but a middle infielder or catcher or CF will gain some value because those are positions where offenses is scarcer, and thus more valuable.
The replacement level adjustment is just a cosmetic change that shifts the scale. Average players have value, so instead of comparing the players to average, they compare them to the theoretical "replacement player," which is the player any team could have for nothing of value anytime they needed one. Your basic AAAA scrub free agent type.
WAR has become a popular "total value" stat because it is listed at fangraphs.com, which is the best site on the web for baseball information (Their pitch data would be fascinating to non-statheads as well. It records things like what percentage of time a player was thrown a fastball, curveball, slider, etc., what percentage of the time they swung at pitches out of the zone, what percentage of the time they made contact on those swings, etc.)
All of the stats can be found in slightly more detailed explanations here: http://www.fangraphs.com/blogs/index.php/glossary/ but it's a little convoluted.
Me again - A little convoluted? Ay yi yi yi yi. Quit screwing with my simple game. That's why I enjoy it, it's simple enough for a simpleton like me to understand.
And I'll throw out this challenge again. You pick a team based on Slide Rule Stats. I'll pick a team based on guys I think, through observation and regular stats (like HR, RBI, average, that kind of thing) can play a little.
I'll kick your ass 10 ways to Sunday, paying no attention to BAPIP, VAPIP, ORP, GORP, VOMIT, GAS, HOSSCRAP and DUNKADOO. The only SABR that will matter is my made-up one: SWS (go through the archives for an explanation - Sir Strasburg has it).
Anybody have any water? The first 10 pills went down easily enough, I need help with rest. Oh, and let's go Caps. Make the old joke (What's Red, White and Blue and plays golf in May? The CAPS!) moot.
7 comments:
Batting average is a pretty convoluted stat, if you think about it, especially compared to something simple like OBP. RBIs tell you more about how good the guys in front of you are at getting on base than how good a hitter you are.
Plus, the more basic rate stats generally are more reliable and conisistant than the "classic" stuff like AVG, R, RBIs. It'll probably make your headache worse, but Needham recently linked to an interesting article from a couple years ago about this stuff:
http://web.archive.org/web/20080102094412/http://mvn.com/mlb-stats/2007/11/14/525600-minutes-how-do-you-measure-a-player-in-a-year/
But I will agree with your ¡Livo! love.
Speaking of Caps, I just voted for him in your 3-week MVP poll, but I was surprised to see that he is getting so little love from your voters. Livo's been great, but he's only pitched in four games. Clippard is my favorite pitcher on the big club right now, but he hasn't been getting those last three outs of each win. Pudge has been hot, but he doesn't even have a home run yet.
Capps has slammed the door on 9 of the 11 Nats victories. He may not have always looked pretty doing it, but he has gotten the job done. Imagine how different this season would be if Hanrahan was still attempting to close.
Two other comments: 1). Don't you think the Cubs wish Capps had chosen them over the Nats? and 2). You can see why the Pirates have had a 20 year run of futility. Releasing a still young player of Capps's obvious ability before he was even eigible for free agency and getting nothing in return was a supremely boneheaded move.
Man, you're as old as you look! :)
One of the brilliant things about Bill James is that he was a writer before a statistician. Most of the statheads on the internet are approaching it the other way.
They don't do a good job of communicating their methods or (most importantly) why they're important.
Remind me sometime, and I'll loan you a copy of one of the old Baseball Abstracts. I can't remember the year off-hand, but he walks you through -- using plain language -- what goes into scoring runs, and why BA stinks on ice.
A simple explanation would be wonderful. I think you would agree the one I posted does not qualify as simple.
And I indeed get older every day. The Nats are trying to keep me young. The Caps are trying to kill me.
"I'll kick your ass 10 ways to Sunday, paying no attention to BAPIP, VAPIP, ORP, GORP, VOMIT, GAS, HOSSCRAP and DUNKADOO"
Well probably not but if those teams played a 10 game series it would most likely end around 5-5.
My problem with the complicated stats is when they are brought in when simpler stats can do the job. Last year Ryan Howard was 16th in OPS. Everyone knows he's not the best fielder or base runner so maybe he's... the 20th best player last year? Ok well what does WAR say? 28th best? Like you say - out of how many players? Do those roughly 8 spots really matter?
There's a place for fancy stats but they shouldn't be the default. Slightly more accurate doesn't make up for much harder to understand.
Well said Harper - mostly I kid on this stuff. I realize they have their place. Too many people have these weapons and don't know how to use them.
Me? I don't have them and don't know how to use them.
Harper and Chris are both correct. Somebody needs to teach the stat-heads to write. And somebody needs to teach them to use the simplest stat necessary that proves your point. If you overwhelm people to prove a simple point, they tune out. If your goal is to educate fans to change the stats they use, confusing them to the point that they tune out is not the most effective strategy. Too often I read a strong point articulated so poorly that there is no way anybody actually understands it. They reach for the advil and move on instead.
Post a Comment