Wednesday, May 23, 2012
The real replacement level of starting pitchingPosted by Derek Ambrosino at 1:26am
I’ve spoken about my high stakes league in which I co-manage a team with a friend. In that league, another two of our friends co-manage a second team. This past weekend, we had a discussion about potential trades in which I may have appeared unreasonable.
What I did isn’t really out of the ordinary; I simply placed an extremely high price on one of my team’s better hitters if I was going to be receiving a starting pitcher in return. Many of us have a sense that pitchers are harder to trade—or fetch less—than hitters. They do. Or, they should.
Most simply, trade value comes down to replacement level. If you play in a daily league with unlimited, or liberally limited, roster moves and are in need of a starting pitcher (as we are), the replacement level starter can actually be pretty high, provided you are willing to exert some effort and display a bit of ingenuity. Elevating the replacement level of a position player is much more difficult.
While one may tempted to look at the top few starters floating on his league’s waiver wire to define a replacement level starter, that’s not the reality for the shrewd owner. Let’s try to estimate what the real value of a replacement starter can easily be with a little effort and planning.
Let’s assume for the sake of this article that 180 innings is a reasonable target to expect from each pitcher in the starting corps. This works out to about 30 innings a month. Remember, nobody says that those innings must all come from the same player. If you can find one opportunity to spot start a waiver wire pitcher per week and fill that spot with a middle reliever the rest of the time, you should be able to piece together a high quality pitching line. (This is easier, operationally, when all roster spots are P instead of RP-, SP-specific; one of oft-stated my pet peeves!)
I’m not a hardcore researcher-writer, so I’m going to pluck numbers here for the purposes of illustration—there are so many variations of league size that calculating the average stats of a waiver wire pitcher would be an exercise in the arbitrary anyway. However, in a 12-team mixed league, it’s not unreasonable to think that your standard omni-available pitcher might have a skill level of something like 4.50 ERA, 1.35 WHIP, six K/9 and one win per 3.5 starts.
You have to find only one opportunity a week when you think one of any of these pitchers, against various opponents, with various skill sets, and in various run environments can perform above his average. It’s not that much of a stretch to think you can coach these pitchers up to a line more along the lines of 4.25 ERA, 1.30 WHIP, 6.5 K/P and one win per three starts. Of course, depending on the choices you make, you could wind up with better rates and lower Ks or the reverse. Frankly, that’s one of the benefits of this approach as well; it allows you to chase what you most need. Though to be fair, if the alternative is trading for a quality pitcher, that player should be able to help you everywhere.
Proceeding, it’s not too difficult to find waiver wire middle relievers who post numbers more along the lines of 2.50 ERA, 1.25 WHIP, 7.5 K/9 and one W per 15 innings pitched.
At the end of the day, you are looking at six innings from your starter per week and three innings from your reliever. This should net you somewhere between 27 and 33 innings per month. For simplicity’s sake, let’s take the mathematical middle and mete out 30 innings of this composition “replacement pitcher” with a ratio of two to one starter to reliever innings and extrapolate that across a six-month season.
That’s not bad for a free additional starter, right? That line doesn’t hurt you anywhere!
So, if I can get this for free, what would it take for me to give up, say a healthy Paul Konerko? A lot more than Tommy Hanson, right? And, if I told you I probably still wouldn’t do it even for David Price, you might think I’m taking it too far, but I really wouldn’t be acting so unreasonably, correct?
These names may or not be random.
My free pitcher may be a 10 or 15 percent drop from a pitcher like Price, with some additional variability regarding wins. For either the star or the composite replacement, win are difficult to predict.
However, when the most attractive first basemen on the wire are players like Justin Smoak, there is just very little chance I can find 85-90 percent of Paul Konerko in the free agent pool. Maybe I can really micromanage meticulously and elevate my replacement level production, but that’s ton more effort than just grabbing one of a dozen relievers and keeping an eye out for decent pitchers facing the Pirates, Mariners or Padres (preferably at Petco).
There’s a second benefit of filling my needed pitching slot this way as opposed to trading a premium bat for a quality arm. My injury risk from that pitching spot is nil. I’m not relying on any single pitcher. I simply need to find roughly four quality spot starting opportunities a month and select from a plentiful pool of capable set-up arms. If I trade for Price and he gets hurt, I gave up Konerko for nothing, or for very little.
The last point may not be totally fair, as I still hold the injury risk of Konerko. So, if he gets hurt and I didn’t trade him for Price, I’m left holding the bag. But, the point stands. Remember, I’m not trading Konerko for Price, but trading the difference between Konerko and his replacement for the difference between Price and his.
One final point regarding this theory: While it may be intuitive to think that you are exposing yourself to a greater degree of expected variance in performance by using a conglomerate of pitchers and therefore not allowing any individual performer the opportunity for his performance to normalize, that is not the case. Your starter-by-committee has a cumulative expected performance as does the individual pitcher. Any greater likelihood of variance would be due to particular pitchers having larger individual variances in expected performance; the fact that you are using one player versus 15 or 30 is essentially irrelevant.
In conclusion, it can be considered reasonable to highly value elite hitting and demand others "overpay” if they are selling you pitching. Of course, if you are hoarding hitting such that you are amassing surplus production that doesn’t influence your point totals, then you need to redistribute your assets. But, remember, it isn’t too difficult to put together a quite serviceable replacement level pitcher from the scraps on the wire, without giving away anything but a floating roster spot.
Derek Ambrosino aspires to one day, like Dan Quisenberry, find a delivery in his flaw, you can send him questions, comments, or suggestions at digglahhh AT yahoo DOT com.