Stealing Ten Wins

After 67 games, the 2012 Pirates are 35-32. Over a 162 game season that is an 85 win pace. However, based on their current run differential, their expected Pythagorean won-loss record for 162 games is 74 wins. What do I make of this? A couple of things:

1. The past only matters as a guide to what is likely to happen in the future. The Pirates actual record is four games above what the run differential would expect. That’s a good place to be in. But they need to take a long look at improving what put them in the position of a negative run differential – a terrible offense.
2. The Pythagorean won-loss calculation is pretty solid. Going back to 2006, about 90% (161 of 180) of the teams were within 6 wins of their expected won-loss record as calculated based on their run differential. That’s 90% of the teams who are within 3.7% (6/162) of their expected record. Nearly 3/4 (131 out of 180 – 72.8%) are within four wins (2.5%) of their expected record. I’d call that calculation an exceptional predictor of outcomes.

So, let’s say the 2012 Pirates are one of the more extreme teams and actually finishes six games better than expected. And lets say, for the sake of argument, that the run differential doesn’t get substantially better (I know, I know, I know – given that we are not even half way through the season, a couple of blow out wins or losses can alter the run differential and make the Pythagorean numbers jump around. Humor me!). What that means is the Pirates expected record based on the run differential after 67 games is the same expected record that they would have based on run differential after 162 games. That would put the Pirates at 80 wins on the year – 74 expected wins plus the six for being a positive outlier (outperforming expectations based on run differential).

We have the new fifth playoff team in 2012. Going back ten seasons, the average number of wins for that as-of-yet-to-be-created fifth playoff spot is 89. For the sake of a round number (and because ‘Stealing Ten Wins’ sounds better than ‘Stealing Nine Wins’), that means the Pirates need to somehow collect ten more wins than they are expected to (74) plus the six for outperforming the run differential in order to make the playoffs. How can this team steal ten wins?

I think this is a terribly tall order. In this excercise, this team needs to make up ten games with less than 100 to go. As it stands now, the Pirates are actually 1.5 games out of the fifth playoff spot. Based on expected record, they are five games off the pace.

Of course, if the first portion of the season leads one to conclude that the team can keep up this actual pace – 85 wins – that would leave the club five games short of 90 wins.

The main question for the front office right now is whether or not they think this team can make the playoffs in 2012. If the answer is ‘yes’, then they will need to devise some ways to swipe those ten wins (or five wins depending on your point of view). Here are some suggestions:

1. Bench or release Clint Barmes. Let Jordy Mercer play the next four weeks. If he doesn’t look good, acquire a young shortstop who is blocked – Paul Janish? – for a reliever and let Mercer and ‘new shortstop X’ battle it out.
2. Demote Kevin Correia from the rotation in favor of Rudy Owens or Jeff Locke
3. Demote Brad Lincoln from the rotation as soon as Jeff Karstens is ready
4. Trade for an impact bat. Preferably a young one. If a team is desperate for arms, package either Locke or Owens and a reliever for a bat. I could be talked into including Robbie Grossman or Starling Marte. Include Correia if the trade partner is willing to take him. Josh Willingham? Seth Smith? Chase Headley? I’m not going to pretend to have a handle on what it takes to get a trade done. Anyone/everyone can speculate about what it would take to land a certain player. The point is simple: the offense is awful and needs to be improved if this club expects to win 90 games in 2012. Changing shortstops is a start, but that won’t be enough.

Footnote: even with what appears to be a good month, Clint Barmes is still having a bad month. His batting average is up in June, but he still ranks 13th among NL shortstops in OPS in June. In his best month he is still very mediocre. His time as a starter needs to end now.

Tags: , , , , , , , ,
  • buccotime57

    Great points…i agree we need to bring up locke or owens and trade the other and maybe a robbie grossman or mel rojas..i still like Jhay and mercer spliting a SS..and i would love to get willingham

  • http://www.facebook.com/craig.biddle.50 Craig Biddle

    Since May 23rd, when they stood at 20-24 after a 3-1 loss to Jon Niese and the Mets, the Pirates are 17-8 for a .680 W-L record. More importantly to Pythagoras, they have scored 117 runs and allowed 90 during that stretch, for a Pythagorean record of 15.7-9.3. If we project THAT Pythagorean record out for the remaining 93 games, we get 58.4 more wins, for an overall total of 95.4-66.6.

    And there’s little reason to believe that this is any worse of a guess as to their expected final record than the current full-season estimate.

    Teams constructed on the model of a couple of very good starters, an excellent bullpen, solid defense, and not so much offense tend to outperform their Pythagorean record anyway, because they do a very good job of winning close and low-scoring games. Like a certain team that is 17-11 in one-run games, but 6-9 in games decided by 5 or more runs. Or like a team that is 3-0 in extra innings, but 34-32 in regulation.

    This is less of an accident than Pythagoras would understand. The Pirates’ net runs differential in the 1-run games is +6, and in the blowouts it’s -16. That’s a net of -10 which is, coincidentally, the same as their overall runs differential coming into tonight’s 4-1 win over the Tigers. To my way of thinking, that means that in the 26 games they have played that finished with a margin of 2-4 runs, they are +3 runs and 14-12 overall. According to Pythagoras, this is an expected record of .514, or 13.4-12.6 or thereabouts in that subset of games.

    Note that their record gets progressively better, justifiably according to Pythagoras, as the margin in the game decreases. And the vast majority (78.2%) of their games have ended with a margin of 4 runs or less. Also note that the Pythagorean estimated wins totally breaks down for one-run games – your runs margin is exactly the same as your net wins – losses, by definition.

    Pythagoras be damned!

    • RandyLinville

      I hear what you are saying and your points regarding the bullpen and winning close games are correct and I believe are essentially the main driver behind outperforming the theory.

      I humbly disagree with your notion that the most recent 25 game spread is equal in validity to the current full season estimate (“And there’s little reason to believe that this is any worse of a guess as to their expected final record than the current full-season estimate”). I think it is unwise to project a full season based on a team’s peak performance.

      Take for example the 2011 team. The peak 25 game run differential was 32 runs (112 for and 90 against) from June 21 to July 22. The Pirates were 18-7 in that stretch and were resting at 51-46 and just a game out of first. If anyone had assumed that the 25 games that carried them to the end of July would simply repeat over and over again were sadly and badly mistaken.

      I don’t expect a collapse in 2012 like what we saw in 2011. But I also don’t expect this team to continue to post 30+ run differentials over every 25 game chunk from now until the end of the season.

      I hope I’m 100% wrong. I hope this team continues to do what it has been doing – all year with the pitching and more recently with the bats. I’m simply not that optimistic.

  • Greatone1210

    Well written with some good points but I hope you won’t mind a few similar and differing thoughts.

    1. Marte should not be included in a trade for someone of either Willingham or Headley’s caliber. Marte is best utilized for a trade that brings us a young slugger with years of control. He has too much potential as a 5-tool kinda guy that is worth more than a slightly above average player. With that being said . . .

    2. Use Marte and another high-ceiling player to grab a young SS like Lawrie or Castro.

    3. If Appel signs, Taillon becomes a trade-chip. I know it is unpopular but imagine the return we could get. It is great we have such deep pitching depth but we need position players as well. You must spend big to build up a strong team.

    4. Correia needs out of the rotation and so does Lincoln. But Lincoln is good in the bullpen so keep him in the organization if possible. Karstens is okay as a back end starter but the future is bright with Cole and the others.

    5. The front office needs to be smart about all this. Try to gain an impact player this year to at least get a winning record. Then continue to build via farm system, draft, and free agency. We must avoid giving up a ton for one year. Build for the future.

  • RandyLinville

    Thanks for the comment and I never mind a differing thought.
    1. I see Marte as more of a fourth outfielder in the R.J. Reynolds mold. A good guy to have on a team, but not a difference maker. Hopefully you are correct and he is an impact player.
    2. Not a huge fan of Castro. He’s better than what we have now, but I think he has peaked as player (like Garry Templeton did 30+ years ago – that is, I don’t see him getting any better. He might sustain this level of production for longer than Templeton did) and too much of his value is in his batting average. He doesn’t walk much and he doesn’t hit for power. That being said, Marte, Locke/Owens and a reliever for him would be fine by me.
    3. That’s a good point. Never thought about it.
    4. Agree with you 100% on Lincoln. I want him on the team, but in the bullpen.
    5. I’m with you on this as well. If they go out and get a bat, it needs to be someone who can impact 2012 and still be a starter on this team for a couple of years beyond this one.

  • http://twitter.com/pskell02 Patrick Kelly

    At some point this season, at least one blogger is going to stop looking at the damn Pythag when talking about this team. I agree and understand that teams generally play to what the Pythag record says they will, but we need to stop worrying about what it says we should be until the end of the season. At this point, Pythag record means diddly squat. The team was historically bad to start the season and are only -10 in run differential right now after tonight’s game. They have shaved 19 runs off of their differential this month. The sticks are starting to come around a little. Barring an implosion of the rotation or the offense dipping back into historically bad levels, they have a very good shot at being even by the end of this month. One more hot month, accompanied by some good trades for bats, I see no reason why this team can’t be 4-5 games over .500 in the season ending Pythag record.

    Also to expand upon the pitching, if we swap out Correia and Lincoln with better options (Karstens, Locke, Owens, Wilson, Vandenhurk), the RA number very well could drop. In turn helping the RS/RA numbers even more.

  • RandyLinville

    That was my point about the Pythagorean – respect it because it is accurate while recognizing that to date the team has been outplaying/outlucking/outwinning expectations based on run differential (and therefore it is somewhat meaningless mid-year). So, if, if, if the FO thinks this team can hang around, we need to fix the issue with the run differential, which is the putrid state of the offense.

  • http://pulse.yahoo.com/_46YTB5OO4O7UM2OVB44UZSTXNA Nate

    If Appel signes, what could we add to Taillon to get Willingham and Sano? I am not a fan of trading Taillon, purely a what if.

  • RandyLinville

    I’d like to think we could get Willingham even up for Taillon. Maybe we throw in a bat, like Clement – someone who could play for Minnesota in 2012 – and a mid-tier, lower minor league bat as well. I don’ t know that Sano would be on the table without the Twinkies being overwhelmed. But, that’s nothing short of wild speculation from me for both scenarios.

  • Greatone1210

    Terrible idea to trade Taillon for Willingham. His best days are behind him and Taillon is worth a lot more than a 33 y/o OF.

  • john.alcorn

    Taillon for Willingham woudl be a massive overpay. No GM in baseball would give up a top 20 prospect in all of MLB for the journeyman OF.

  • RandyLinville

    I think you are right and I’m not advocating such a swap. But since the ‘what if’ question was asked, I’d like to think that it would be a one for one trade if both teams got together to talk.

  • Chauncey Jordan

    Your credibility really takes a hit when you suggest we would consider swapping Taillon for someone like Willingham.
    Otherwise, an interesting take.
    I think the offense is certainly improving upon the first 40 games.

    They are not trading Taillon. No chance

  • RandyLinville

    I agree – it would be an overpay. I’m not advocating such a move. The question was asked of me and I answered it. Simple as that.

  • RandyLinville

    If you read the comment below, the question was asked of me ‘what do you think we’d need to add to Taillon to land Willingham’. It was a what if. I answered a what if. I’m not advocating trading Taillon for Willingham.