Data-For-All

Thank God somebody has come along to take a closer look at BABIP. Far, far too many people take it on face value as something completely out of the pitcher’s (and to a lessor extent, a batter’s) control. Worse, it seems like every fantasy baseball “expert” uses it as some golden value that can be used to discredit pitchers (and batters) having a good season or proof of somebody having a bad season has been just unlucky instead of just bad. Intuition and common sense should tell you this isn’t true. And if you lack these things, just watch Jeff Weaver pitch sometime how he grooves pitches once he’s lost his focus and the increased number of line drive hits (which typically are correlated with increased BABIP figures), what with his career 22.2% line drive average against him and his career 0.308 BABIP against figure.

Anyway, you can find a link to Keith Isley’s article by clicking here.

Which is harder to believe?

A) That I drafted J.J. Hardy (you know, the guy leading the NL in both HR and RBI … as a SS!) in the 27th round of a deep, 12-team mixed league? He was the 314th pick out of 324 players taken.

B) That I dropped him right after the draft for “slugger” Luke Scott, OF (0.336-10-37 in just 214 AB last season as a 2nd half callup) … now batting 0.236 and losing playing time (… and also now off my team)

In my defense, I already Miguel Tejada as my SS (from my 2nd round pick), who I was happy to land. However, as it turns out, Hardy is taking the world by storm (he just hit another one out today - he can’t be stopped!), Scott has long been since dropped from my roster, and Tejada’s power seems to have evaporated over the past year and a half (2hr and 0.383 SLG% so far).

I guess I can’t complain too much about my “bad luck” with letting Hardy slip away, though, since I’ve gotten lucky with bounceback seasons by T.Helton, B.Bonds, and J.Peavy, which I got later than they should have went.

In a previous post I created a decision tree using R to identify what spring training ‘06 stats could be used to best differentiate who went on to have a good or bad April ‘06 (among the sample of players in Ryan Armbrust’s study of Do Spring Training Stats Mean Anything?).

To recap, for the players in his study (p.s. all players who had at least 50 at-bats in Spring Training of March 2006 and accumulated another 50 in the regular season’s first month of April) I suggested the following logic could be used to classify high April ‘06 OPS vs low April ‘06:

• If player hit 5+ hr in spring training ‘06, then good April ‘06 was seen (avg OPS=0.933)
• But, if player hit 4 or fewer hr and had a BB/SO ratio of less than 0.2361, then a bad April ‘06 was seen (avg OPS=0.631)

So, naturally, I’m curious as to how well this “model” based on last year’s data remained true or not for this year.

Read the rest of this entry »

It’s a crime that Ian Kinsler (0.333/.405/.739) with 8 hr, 20 rbi, and 4 sb so far this season through 20 games - the only guy on the entire Texas Rangers team batting over 0.300 and more hr and rbi than the #3 hitter, M.Young, and #4 hitter, M.Texeira combined! If there’s any consolation here it’s that he’s finally been moved up to #7 against RH instead of being stuck batting 9th every game, but still. This has been ridiculous. It makes you wonder how much of the blame for their 8-13 start is directly attributable to the manager not putting the best lineup out there and instead going by players’ past performance rather than their current performance.

Ryan Armbrust has an extremely interesting article “Do Spring Training Stats Mean Anything?” at www.beyondtheboxscore.com that looks at correlations between ‘06 spring training stats and performance in ‘06 April. To help keep things clean, he looked only at players who had at least 50 at-bats in Spring Training of March 2006 and who also accumulated another 50 at-bats in the regular season’s first month of April. There were 135 such players, which seems like a pretty decent sample.

His findings are that the correlations between spring training ‘06 and April ‘06 for AVG, OBP, SLG, and OPS are all pretty weak and practically insignificant. His reasoning for why the weak correlations is solid, I think. Basically, the population of players is different. For one thing, the pitching talent is diluted. Batters face a lot of pitchers that will not even stay at the major league level this year, other pitchers are tinkering with new types of pitches, and batters themselves might experiment with different mechanics as well.

As an aside, somebody in the comments after the article poses a really good question; namely, what are the correlations for these metrics from month-to-month during the season? Are they any better? It seems pretty clear that spring training stats (as a whole) don’t really predict April regular season stats very well, but do April stats predict May stats any better? What about August to September? It would be interesting to see the correlations from month-to-month.

Read the rest of this entry »

www.baseballmusings.com has a nifty tool to compare all players between two dates. For example, look at all the players for the Cubs in June 1984 (min 40 PA), sorted by batting average in descending order here or look at all players across all teams for the same criteria here. Pretty cool!

Check out this cool website, http://www.nationmaster.com

Wow! The US is only 16th in the world in “Broadband access (per capita) by country” and 8th in “Happiness level = Very happy by country“, among other statistics.

Today I thought I would try doing something interesting and fun using SQL 2005 as a way to nudge me to get started learning it since, at some point in the next year, we’re going to want to start moving towards it for new apps. Obviously, it’s no fun to just do something you can already do with a previous version, so I thought I would dive in and learn how to use the new .NET CLR feature, which gives you the ability to develop stored procedures using .NET rather than native T-SQL.

Read the rest of this entry »

Ever since upgrading a Windows 2000 workstation to Windows XP about a year ago, the Scheduled Tasks folder on \\SaidWorkstation stopped appearing in Windows Explorer. After checking permissions everywhere, shares, comparing registry values to other workstatons where Scheduled Tasks folder does appear, and pulling out my hair, I noticed that I could not do a remote registry connect to the bad workstation. I received the following error:

Cannot open HKEY_LOCAL_MACHINE: Error while opening key.

I then searched on this error and found KB892192, which I tried and, voila, the Scheduled Tasks folder came back to life!

Now, what this has to do with Scheduled Tasks is beyond me. But, ironically, it was a permission thing afterall, albeit a convoluted and indirect one. So, I’ll be praying for anybody experiencing this issue to find this article and save yourself about 50 hours of frustration.

I haven’t been blogging much lately, but I did update my blog anyway because I was sick of clearing out 50+ spam comments  per day - which is ironic considering I barely get 50 legitimate comments a year! So, now I require an anti-spam image be typed in. But, don’t worry, it’s not one of those impossible to read ones. What’s scary is I’ve read spammers are already finding ways to read them!