Fangraphs: Twins Team Win Values
Over at Fangraphs, Dave Cameron (co-founder of USSMariner.com) talks about the success of the '08 Twins in producing runs. It's nothing new to us, but it shows us exactly where Minnesota sat in situational hitting versus the league average.
about 1 year ago
Jesse
10 comments
0 recs |
Email
PrintComments
Past returns are no guarantee of future results
>>While watching your team capitalize on a huge portion of their run scoring opportunities is exciting, it’s not a great recipe for success.
Of course, it is a great recipe for success. The Twins won 90 games, nearly, and that’s success. If your team hits well in the clutch, it will probably be successful.
No way to tell if what he’s intending to say, but failing at, is that hitting well with RISP is not an established way to sustain success. It probably is what he’s trying to say. Of course, having a good team ERA one season is no guarantee of sustaining success, either.
How hitting with RISP is less of a certainty from year to year than any other stat, such as staff ERA, SOs, BBs or team SB, BA, RBI or SA is puzzling.
Bloggin' the bloggers since 1938.
by Johnny Safron on Jan 27, 2009 7:34 PM EST reply actions 0 recs
Hitting with RISP is less of a certainty from year-to-year...
…for many of the same reasons that hitting in the month of April is less of a certainty from year-to-year. Players get fewer chances to hit with RISP, so we get less data to form a conclusion on. Also, there is a problem in that not all RISP situations are the same—the league as a whole has a much higher batting average with RISP and less than two outs (nearly .300), but the league as a whole has a much lower batting average with RISP and two outs. So not all RISP situations are created equally and your set of opportunities might have a bigger impact on your average with RISP than your situational ability does.
Personally, my belief is that situational hitters exist, but that traditional statistics like batting average with RISP are insufficient to determine who those hitters are. Also, while situational hitters exist, for 95% (or more) hitters, it’s not really a big deal, or the statistics that we do have wouldn’t work as well as they do.
The smart money is on the Twins not hitting as well with RISP next year, because very few teams in baseball history have done as well as they did last year, and almost all historic performances are a combination of skill and luck. The Twins still have a lot of good contact hitters on the roster, though, so I think they are a good bet to have a better-than-average batting average with RISP next year.
by ubelmann on Jan 27, 2009 10:36 PM EST up reply actions 0 recs
What is a certainty from year to year?
Nothing.
For example, no one before the 2008 season would have expected a part-time, 27-year-old third baseman with a couple dozen big-league games under his belt to, in 218 ABs and with a 2-1 SO-BB ratio, contribute 72 runs (Buscher: 47 RBI+29 RS-4 HR) in 62 starts.
And that alone suggests that Twins will falter in this area, as he also hit .386 with RISP.
Considering everyone wants him replaced and it’s almost impossible for a player ot replicate that production, it’s almost a given the team RISP will fall, and it’s also going to be difficult for the 2009 third basema(e)n to produce the number of runs produced by Twins’ third basemen in 2009.
Those who want Orlando Hudson signed are probably right to address that, rather than third base.
But back to the lack of certainties.
There was no certainty the Twins would be successful on nearly 80 percent of their stolen base attempts in 2008, as they were in 2007, and they weren’t, despite the addition of speedster Carlos Gomez. I suspect it would be easy to scour the Internet and find those who thought the Twins would be more productive or as productive in basestealing in 2008 compared to 2007, but an 80 percent SB rate is so high – just as the team’s 2008 RISP BA was – that it was a good bet this figure would not rise, and probably would fall.
It would not be difficult to find forecasts of a “certainty” that, given a healthy Mauer and Morneau and a decent amount of PAs from Kubel, that the 2008 Twins would hit more home runs than they did in 2007, particularly with the addition of Delmon Young. Yet Kubel had more than 500 PAs – a career high – and Mauer and Morneau each played in a career-high number of games, and they did not hit more home runs in 2008.
The 2008 Twins lost Johan Santana, added every fan’s most hated starter in Livan Hernandez to a young pitching staff and turned in essentially the same team ERA as in 2007. The “smart money” was probably not on that happening.
Of course, none of that changes the notion that hitting well in the clutch is indeed a recipe for success, contrary to Dave Cameron’s assertion. I really can’t see how hitting well with RISP is a recipe for failure, do you?
Bloggin' the bloggers since 1938.
by Johnny Safron on Jan 27, 2009 11:16 PM EST reply actions 0 recs
Come on
Just because these things are 100% certain doesn’t mean that there aren’t different levels of certainty. Just because I don’t know with 100% certainty what the weather is going to be tomorrow doesn’t mean I can’t trust the 1-day forecast more than the 3-day forecast.
If you look at Cameron’s entire last paragraph, the assertion is clearly in the context of future performance, and that past overall stats are more indicative of future overall stats than past RISP stats are indicative of future RISP stats.
by ubelmann on Jan 27, 2009 11:23 PM EST up reply actions 0 recs
Well phrased!
That would have taken me twice as much space to state.
by PhoenixV on Jan 28, 2009 3:55 PM EST up reply actions 0 recs
Once again...
I really can’t see how hitting well with RISP is a recipe for failure, do you?
Bloggin' the bloggers since 1938.
by Johnny Safron on Jan 28, 2009 9:09 PM EST reply actions 0 recs
There are two questions here
1. How much of our performance (relative to WAR) was a result of hitting with RISP?
2. Can a high RISP average be repeated from year to year.
On #1, my guess is that about half of the ~8 wins Cameron cites can be allocated to RISP. The other half appears to be baserunning and doing the “little things” that are not captured in wOBA.
On #2, I’m not aware of any evidence that this is repeatable by a given team from year to year. PhoenixV had a very good analysis of our chances to regress to the mean.
by Adam Peterson on Jan 28, 2009 10:19 PM EST up reply actions 0 recs
Probabilities
You keep talking about certainty. In that I agree with you: Outside of mathematical theorems, nothing is certain. Ad they’re so abstract as to be inapplicable for practical science. For that we use probabilities. This might seem incredibly basic, but your straw man that we are claiming certainty for predictions requires a basic approach. So here goes.
Some things are more likely than others. For example, the Twins average with RISP last year was an unlikely event. In all likelihood, it won’t happen again. The most likely event is that it reverts to the average the team has put up since Mauer and Morneau started hitting in the middle of the order, something like .290. It could revert to other years, like .260. Or it could continue to defy the odds and stay above .300. But the odds of either extreme in the continuum of possibilities happening are pretty low, maybe 50 to 1, off the top of my head.
And projection is like that. We can’t be certain of our view of the past, and we’re far less certain of our view of the future. We just make educated guesses based on probabilities, comparing players with past players of similar ages and track records. It’s not even close to being certain. But if you balance three or four projections for a player, you’re more likely to come close to what he will do in the coming year.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
by cmathewson on Jan 28, 2009 10:27 PM EST up reply actions 0 recs
That was essentially covered in the original post
>No way to tell if what he’s intending to say, but failing at, is that hitting well with RISP is not an established way to sustain success. It probably is what he’s trying to say.<
Which does not change the fact that if you hit well with RISP, you will likely be successful.
It’s amusing that numbers-obsessed fans of other teams have eagerly latched onto this stat, taking comfort in the notion that a likely drop in that number forecasts failure. Hitting well with RISP will lead to runs. So will hitting a lot of two-run home runs with a runner on first. A team’s success with RISP can drop, yet that team could produce more offense.
And hitting well with RISP when you’re ahead 8-3 is less of an ingredient for success than doing it when you’re behind 8-3.
As with most numbers, it’s not that they happen, it’s when they happen.
Bloggin' the bloggers since 1938.
by Johnny Safron on Jan 30, 2009 1:06 AM EST up reply actions 0 recs
gambling
In poker, getting a full house is a great recipe for sucess. However, if I have a succesful round of poker, and I do it by getting a lot of full houses, one would expect that I would do less well, on average, in the future, since I am unlikely to replicate getting good cards.
Conversely, if I have a successful round of poker, despite getting junk hands, it is a good indication that I might do even better in my next game.
If you can consistently maintain high RISP, then you have a great chance at success. If RISP mostly luck, then succuess with high RISP indicates that you will perform less well in the future. The opposite is also then true.
by snolls on Jan 29, 2009 9:27 AM EST reply actions 0 recs
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2010 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
