With the economy sapping the Canadian dollar, we knew the Toronto Blue Jays wouldn't be quite as big of a player this off-season as they may have hoped. Dashed were hopes of landing a DH like Milton Bradley or re-signing A.J. Burnett. Instead the Jays are looking for value pick-ups, something they've done in the past with decent success.
Just last off-season the Jays signed Matt Stairs, acquired Marco Scutaro (for Graham Godfrey and Kristian Bell), signed Shawn Camp, Shannon Stewart, Armando Benitez, and signed David Eckstein. The Jays would later turn Stairs and Eckstein into Chad Beck and Fabio Castro. Meanwhile Camp turned out to be a decent reliever and Scutaro was perhaps the biggest surprise of 2008.
Thus far this off-season the Jays have plucked oft-injured starter Matt Clement off the heap and apparently have some interest in Carl Pavano. The "Whys?" of these moves are obvious:
The Jays need starting pitching to replace their own injured starters in Dustin McGowan (hopeful to return around May) and Shaun Marcum (out for most of 2009) as well as replacing Burnett. Currently starting options include Clement, David Purcey, Jesse Litsch, Scott Richmond, and a number of Triple-A starters.
The Pavanos and Clements of the market are undervalued. Teams are unable to rely on injured starters for entire seasons, which is irritating and annoying, but good pitching for fewer innings is usually better than league average pitcher over more innings.
That last point doesn't get enough credit, but it's true. Clement's career FIP if 4.24 and Pavano's 4.21. Over the same amount of innings either will save one more run than an average American League Starting Pitcher (FIP of 4.48). If Clement could give the Jays 80 innings he'll save 13.2 runs, Mr. League Average would save 14.1 in 100. That's not a very wide margin despite an extra 20 innings, and the Jays other starters aren't guarantees to be league average anyways.
I suppose this point is less about the Jays, who have starting pitching depth capable of being above league average, and more about the stigma associated with injured starting pitchers. In order to figure out how many innings each rotation slot should be expected to pitch, I took the top 180 IP qualifiers (figuring teams use at least six starters) and separated them into tiers of 30. Here were the results:
Number One: 215 IP
Number Two: 193 IP
Number Three: 172 IP
Number Four: 137 IP
Number Five: 93 IP
Number Six: 56 IP
Of course some teams use more (or less, rarely) than six starters, but that adds ups to an average of 5.3 IP per game. There aren't many teams that would rely on Clement as anything higher than a three starter. Meaning we can focus on Clement as a fourth or fifth starter.
As a four:
| IP% | xIP | NLRS |
| 85 | 117 | 19.6 |
| 75 | 103 | 17.3 |
| 65 | 89 | 14.9 |
| 55 | 75 | 12.6 |
As a five:
| IP% | xIP | NLRS |
| 85 | 79 | 13.3 |
| 75 | 70 | 11.7 |
| 65 | 61 | 10.2 |
| 55 | 52 | 8.7 |
Replacement Level Starter + Clement v. Average Starter
| Four |
Clement | RPLSP | ALSP | Delta |
| xIP | 117 | 20 | 137 | 0.3 |
| 103 | 34 | 137 | -2 | |
| 89 | 48 | 137 | -4.4 | |
| 75 | 62 | 137 | -6.7 | |
| Five |
Clement | RPLSP | ALSP | Delta |
| xIP | 79 | 14 | 93 | 0.2 |
| 70 | 23 | 93 | -1.4 | |
| 61 | 32 | 93 | -2.9 | |
| 52 | 41 | 93 | -4.4 |
So what does all of this mean? The difference between Matt Clement and a Replacement Level Starter as a four (or five) and a league average starter is less than a win. Now there are some obvious points worth mentioning:
There's no guarantee Clement pitches to his career FIP.
The pitcher replacing Clement is probably better than replacement level.
Which is why I don't want this to be an analysis of Clement only, but rather a thinking point towards all injured starters, especially those who have shown the ability to pitch near 100 innings. It's not ideal, but if you can get pitchers capable of that type of performance for 800k, why wouldn't you take the chance?