Sabermetrics, just a number crunching exercise or great value to baseball management?
MCMLVTopps
Posts: 4,618 ✭✭✭✭✭
When I've drilled down a bit into Sabermetrics, which is really an acronym for Society for American Baseball Research, which was founded in 1971, I see a bunch of decimals that I wonder how much meaning they really have. If you're a statistician, or just really love playing with numbers, then this is more than likely an enjoyable enterprise for you to review and ponder.
In general, this exercise is generally used to appraise the performance of a player being considered for a team, end of season performance, or used to analyze a team's performance over time. While the numbers and decimals can be impressive, they don't give any indication as to how a player or team will perform in the future, it all relates to previous and current performance, all data of which is "analyzed" by some software that spits out a number. Taking RBIs as an example...while it might be impressive to see that a player had 3 RBIs in a game, or had 80 or 90 in a season, it only indicates a past performance. The player can only excel in this statistic if he has runners on the bases, particularly on 2nd or 3rd, and the batter gets a hit to drive in the run(s). The player has no control over previous batters getting on base, therefore, this stat in terms of Sabermetrics, I think, has little to no value regarding future performance.
Some of the Sabermetric measurements are"WAR Wins Above Replacement, an odd name for the measurement used to quantify a player's overall contribution to the team's wins. Two different versions to this, each use a different formula.
OPS, this is a measurement of a batter's offensive output.
WOBA, confusing to me.
BABIP, Batting Average on Balls In Play. Some luck involved here I think.
ISO, this subtacts the player's batting average from his from his slugging percentage.
UZR, Ultimate Zone Rating. Too deep for me.
FIP, Fielding Independent Pitching.
SIERA, Skill-Interactive ERA.
All this data must fill a laptop with colorful graphics and never-ending updating of stats. It means nothing to me, but I'm sure it has value to some people who stare at this stuff like a stock market software program. These acronyms were created by Bill James, in 1980, who claimed Sabermetrics "is the search for objective knowledge about baseball". I wonder if he didn't just create all these stats to just look at numbers.
Whenever I watch baseball games, I surely don't care too much about a player's BABIP, his UZR, nor his SIERRA! For those who enjoy such things, I hope your favorites players FIPs, WOBAs and BABIPs continue to excel. 
Comments
Al, your way over my head on all that stuff.
I think this is an arbitration tool, both used by the player and team sides.
The team side finds the negative ratios while the player side finds the positive.
I have no clue either what, or how Sabermetrics is viable to use for "the search for objective knowledge about baseball".
How can it be "objective", when it pertains to a past performance trying to predict a future result or outcome?? It simply can't be objective with so many variables are involved, like the RBI stat. If the batter doesn't have runners in front of him, he doesn't get, or improve his RBI statistic.
Maybe so, but it just seems so convoluted to me. Playing with enough statistics in order to generate all the acronyms highlighted above. Too many unknowns and variables, from injuries to bad luck to slumps to playing so badly the player can be great for a little while, then DFA'd, or sent down.
when I think of stats from my childhood.
Only stat I need.
“ If he’s a good hitter why doesn’t he hit good. “
"I spent 50% of my money on alcohol, women, and gambling. The other half I wasted.
Same here
I’m sure these geek stat systems have value but I honestly never could wrap my head around that stuff.
I’ve always been a BA, HR and RBI guy lol
Sabermetrics is basically common sense put down on paper, and you just gave a common sense sabermetic analysis of RBI and why RBI is not really a good measurement to make an evaluation on. One can make RBI a little better measurement by accounting for the number of baserunners, but then that is still just half the of the equation to offense, as it ignores the getting on base portion of hitting, which ironically is something that traditional fans ignore(ability to get on base), even though they love RBI....but RBI only really come with guys 'getting on base'...which again is quite ironic.
Not really going to go into detail on each of the advanced stats and which are better or not. Simple OB% and SLG% get 95% of the job done and neither are very complicated, and when put together they are extremely accurate to measure the ability or contribution of a player.
Predicting the future is a different subject. However, what typically happens is that a fan will dismiss a sabermetric stat under this premise, but then will use a stat like RBI or batting average to do exactly what they are dismissing.
You can predict the future pretty good in baseball for players and teams, based on what they established as an ability to produce in MLB. A stat like RBI will produce a lot of lies like this for you(as you laid out). A stat like batting average will lead to many false overvaluing of players, although batting average is fairly predictable for future performance in that area of measurement.
I say "fairly" predictable because there is a lot of luck involved in batting average too and that is why BAPIP(Batting Average on Balls in PLay) is used because it does show when a player has a higher batting average than what his skills are showing....basically showing that his batted balls just happen to be finding the holes more luckily than others players batted balls are. When someone has an extremely high BAPIP it is almost a certainty that their current batting average will drop, and it does when given enough at bats for the luck of balls finding holes or gloves to even out and result in a batting average more in line with his true ability.
In simple terms, this is why you don't award batting titles in May
Scouts have the hardest job because they have to predict who will be a star in MLB based on the tools and makeup of a player. They do not have any MLB past performance to rely on and that is why you see so many first round busts in MLB that never even make it out of the minor leagues because that type of 'eyes' evaluation is extremely subjective, faulty, and often flat out wrong.
That is why past MLB performance is such a valuable tool as opposed to 'eyes'.
When you have 2,000 MLB at bats to judge how good a player is, then you have a much better chance of knowing what the player can do for the next couple years(with luck/chance always tugging both ways to muddle that final number each year).
That is also why you see pitchers being evaluated mostly on their K/BB/HR allowed because those are the things most in their control and independent of all the luck/chance that occurs when a batted ball is hit somewhere within the lines and walls. In fact, using a simple K/BB ratio has been extremely valuable in predicting pitchers who will have good seasons the following year because you are isolating their ability as opposed to the fielder's ability that sometimes shows up in their ERA. How much the fielding team is responsible for a pitchers' ERA is debatable.
Age regression, injury, or someone who signed a big deal and just doesn't care anymore, are all things that will make future predictions hazardous, but in the end, the measurements such as OPS+ are going to give damn good year to year predictions going forward. Only the bad GM's will use stats such as Pitcher Wins or RBI as they have all the pitfulls of future predictions of sabermetric measurments but also that those stats have a low level of validity to begin with.
Yes, I am fully aware of the minor tics and tics of even the most advanced and valid measurements and I often point those out myself.