Realism or Reality


Strat-o-matic is often hailed as one of the more realistic baseball simulation games.  As I looked at the standings in my Strat Leagues - yes, I'm now in two of them - I was struck by the irony that both of my teams were in first place in their respective divisions, but both had been outscored by their opposition.  Those familiar with Bill James Pythagorean Theorem would say that both squads have probably overachieved to this point and are destined for a tumble.  After all it's rare that a team posts a winning record despite being outscored. 

However, it does happen in reality.  Three teams have been outscored yet posted a winning percentage of .550 or better: the 1932 Pirates, the 1984 Mets and the 1997 Giants.  The 1987 Twins won the World Series despite being outscored 806-786 during the regular season.  Some would insist the Twins were lucky that year because they played in a weak division and they only had to win two 7-game series to win it all.  Yet they were 43-41 versus the East, largely by beating up on the same teams - Cleveland, Milwaukee and Baltimore, against whom they were 25-11 - on which their AL East counterparts feasted.  They were 24-21 in one-run games, so it's not as though they were overly lucky in games where luck is a significant deciding factor.  Some would suggest that it was their decided advantage at home, where they went 56-25 that year, as the most significant factor.  Yet they defeated Detroit in the ALCS in five games with three of them taking place in Detroit.  Could the fact that they had the best defense in the majors that year have had a large part in their success?  That's hard to say for certain as well since current defensive measures still leave a lot to the imagination.  I would say that we don't really know why they won; just that they did.  Maybe they were actually a better team than their 1987 statistics indicated since the following year with much the same personnel they won 91 games and outscored their opponents by 87 runs.  Oddly enough, they finished second in their division in 1988.  Were the 1987 Twins lucky or unlucky?  More often than not the statistical analysis coincides with what happens on the field but it really only describes probabilities.  Anomalies such as the 1987 Twins just don't reveal themselves easily using such conventional statistical tools.  

Back to the original focus - is my Strat team (the Montreal Funiculars) due for a fall?  I built it with the idea of holding the opposition close with excellent defense and decent starting pitching but winning in the later innings with a bullpen that should shut the opposition scoring down on most nights.  However, my 2004 version of Eddie Guardado (who has allows no baserunners to lefties on his card) has an ERA of 9.25 and my 2004 Brendan Donnelly is allowing 1.62 baserunners per inning.  My automatic 2004 Eric Gagne - the real one posted an ERA of of 2.19 and a WHIP of 0.911 - has an ERA of 2.70 and a WHIP of 1.200.  So has my team been lucky to get as many wins as it has, or has it actually under-performed because the bullpen has been far worse than anticipated?  And why is this relevant? 

Well, the Washington Nationals haven't inspired much confidence that they will last the summer despite being in first place by 4 games.  The reason is that they have been outscored, the result of an apparently lousy offense.  Yet like the 1987 Twins and the Funiculars, they have excellent team defense, plus a pretty solid bullpen.  Another thing they have in their favor is that they are not likely to continue to be outscored.  While they are second to last in run scoring, their team OPS is 22nd overall.  Isolating their production on the road away from pitcher-friendly RFK, they are 15th overall in run scoring, 10th in OPS.  And this has been with Cristian Guzman barely hitting his weight, and without Jose Vidro who should return sometime around the All-Star Break.  So are the Nats lucky to be winning, or have they been unlucky they haven't scored more?  Probability says that if they continue along the same path their record will come back to the mean.  But factors like quality defense and bullpen - both of which the Nats have - can keep the parade going until the reinforcements arrive. 
 
I looked through the play-by-play (the closest I can get to actually watching the games in Strat) to see if I could identify what was ailing the Funiculars and a couple of things stood out.

The first was that HAL, the computer manager, had a nasty tendency to pinch-run for Mike Lowell in the later innings.  As part of my manager strategy, I explicitly forbid HAL from pinch-hitting for Lowell under any circumstance.  But since there is no similar setting for pinch running, it has decided that any player is fair game for that.  Lowell gets targeted because he is the team's slowest runner.  Strangely enough, despite an AA steal rating which almost guarantees success and numerous baserunning opportunities granted by HAL's pinch running scheme, Scott Podsednik (my designated pinch runner) has stolen only one base.  Apparently it doesn't like to green light pinch runners to steal; it just likes to find ways to take the team's best glovemen out of the line-up.  But HAL doesn't stop there.  Rather than a fairly easy position switch where ARod moves to third and Pokey Reese comes in to play short - a move which does not sacrifice any team defense - it opts to put the team's worst possible defensive option, BJ Upton, at third for Lowell.  While it may seem like I'm jumping to conclusions about HAL's integrity, in my other league it quite frequently uses a centerfield-only player as a replacement at shortstop rather than the two players who are eligible to play there.  Regardless, HAL had to be stopped.  So I dropped Pokey Reese in favor of Alfredo Amezaga because the latter could play all the infield position with some expertise, while Reese was limited to just short and second.  Then by specifying Amezaga as the designated defensive replacement everywhere in the infield, there's less chance that HAL will conjure up another crazy defensive scheme.

But because of HAL's strategic lapses, the bullpen was allowing more baserunners than it should, either by hit or error, which meant they stayed in games longer.  That in turn, caused them to be overused with the end result being a loss of effectiveness.  So I had to add another bullpen arm to spread the workload before the entire staff collapsed from exhaustion.  The problem there was that another player wasn't in the budget.  Someone had to be dropped.  That someone was Miguel Olivo, my lefty-killing (.978 OPS against in 2004) catcher.  The decision was made easier because a) he wasn't hitting lefties... well, to be honest he wasn't hitting anyone with his .115 batting average, .148 on base and .269 slugging, and b) his salary was substantial enough to allow for more than one player move.  So I replaced him with the much cheaper (and hopefully more productive) Toby Hall and added Jay Witasick to the bullpen.  Witasick doesn't have a great reputation, but if his card reflects the numbers he produced in 2004 against right-handers, he should be more than adequate as a situational reliever.  The trade-off also left some cap room to add another player should it become necessary. 

Just like the Washington Nationals, despite what the numbers say right now, they should in theory play more toward their record than their run differential as the season progresses.  At least, that's what this fan hopes.