Passionately undeveloped opinions on the state of baseball, the shifting landscape of stats and analysis, and the opiate power of El Pato tomato sauce
Category Archives: Interleague Update
We are officially past the halfway point of interleague games in 2013 (2 games past technically) so I’m back with an update on how the scoring has broken down. First and foremost, the AL leads the NL on record, 80-72. Based on pythagorean expectation, the AL should’ve won 80.82 games, the NL 71.18, so we are actually very close to what our run differential suggests. Check out how win expectancy has tracked with actual wins here:
Not too much variation off expected, in fact never more than about 3.5 games.
Run differential itself is a bit more interesting. The AL regained the advantage around game #30 and has never looked back, pushing it as high as 60 in their favor.
Who wants to see stabilized data? Everyone! Runs scored per game has tracked nicely since about game #40, as can be seen below. There’s currently a 0.29 runs/game advantage for the AL (AL: 4.32, NL: 4.03), though both are ever so slightly below their intraleague averages (AL: 4.39, NL: 4.06). I’m going to step out on a limb and assume those aren’t significant differences.
Lastly, I’ve added a histogram of win margin to see if one league or another is posting a lot more 8-run victories than normal. Nearly 75% of both teams wins are of the 1 to 3 run variety, though the AL has more of the 2- and 3-run variety, the NL has more 1-run. Beyond that each league is posting about the same number of blowouts, with the AL having two 9-run victories to the NL’s zero, and two 10-run victories to the NL’s one being the biggest contributors to run differential.
As a whole, the leagues are very well matched up through a 152 games, which should be expected. The AL seems to be making slow but steady progress towards a better record, but the small win margins would seem to indicate it’s fairly luck driven and not necessarily an indicator of a real quality difference between the leagues. I guess the All-Star Game will sort that one out for us! /sarcasm/
And boy how things have changed. I suppose it’s not actually too much considering one week of data is so small, but still things are different looking. That primarily is due to the White Sox getting swept by the Nationals a week and a half ago, as well as the Twins getting knocked around by the Mets in between snow storms. The AL gained some ground back with wins by Texas, New York, and Baltimore, but really only enough to stem the tide for the time being.
As you can see, the run differential took a drastic turn south in favor of the NL when the Metropolitans beat the Twins by 11. It peaked as high as +16 in favor of the senior circuit and hasn’t got closer than +8. Note: The Twins now play the Marlins for the first interleague series this week. God knows what this graph will look like come Thursday.
Hand in hand with getting outscored is losing (learned that from Joe Morgan), and you can see a nice 5 game losing streak right after I started this project. The AL Central blows. It did provide us with our first drop in expected win percentage though (at game 10). A nice recovery towards expectations has the AL being only slightly unlucky according to Pythagoras.
Meanwhile, the NL has a nice steady climb going for it, which serves up some confirmation bias for me that the AL is top-heavy in good teams while the NL has more parity but higher average talent. Provable? Hopefully for someone else.
And lastly, a new graph that shows each league’s runs scored per game in interleague contests. The NL has actually scored exactly 5 runs per game to this point, whereas the AL is a tick below 4.5. We’ll see how much the summer months cause this graph to go skyward.
I know I mentioned that Ducky wasn’t as positive this data is actually meaningful a few weeks ago and I’m still not either, but I do think it provides an interesting storyline to how the first year of year-round interleague progresses. Who’s the better league is always a nice subplot to the season and while this in no way provides an answer to that question it does provide food for thought. And pretty graphs.
Update: I think Ducky is not as enthused about this as I am, and he brings up good points. I’ve always seen pythagorean expectation used for a single team over the course of a season and assumed it would apply for a league as well. That may not be the case. So while I do some digging take all this with a grain of salt. Which you should anyway since 6 games doesn’t not a meaningful sample size make. The run differential trend should be fun regardless though. -High Pockets
Update 2: Also, this expectation equation I’ve used has a factor of 2 involved. Apparently research since the original equation came about has found that a factor of 1.83 is more accurate. It will be used from now on (if pythagorean expectation is continued).
Since Interleague play is constant this year as a result of the Astros moving to the AL, we can follow how each league is doing all year long instead of waiting until June. I’ll be posting a few graphs after every week to get an idea of what’s going on. The first will be run differential, then a graph of AL Wins vs. Expected wins (computed based of simple pythagorean expectation), and finally a graph of NL Wins vs. Expected Wins (the inverse of the AL graph).
So there you have it. After one week the AL is clearly superior to the NL /incorrect assertion/. See you all next week!