Braves Extension Analysis: Kris Medlen

This is the fourth in my series on possible extensions for the Braves' young core players. The first post covered Martin Prado, the second looked at Jason Heyward, and the third analyzed Freddie Freeman. In the last post, by the way, about 2/3 of TC readers favored extending Freeman at the cost I proposed ($42M over 6 years).

Some contracts are particularly risky. Those given to pitchers are the top example, thanks to unpredictability of arm injuries. Then there's age: the older a player gets, the bigger a risk he becomes. Players with limited track records tend to be more risky, too.

Scarily, all three of those risks apply to the subject of this post, Kris Medlen. Medlen is 27 (and thus perhaps already past his peak), has only thrown 316 MLB innings, and has already had one major arm surgery.

I wouldn't be writing this, however, if I didn't think there was a case to be made for extending Medlen. We can dismiss his obviously unsustainable ERA from 2012, but his peripherals were still fantastic (his 2.22 FIP as a starter is a half-run lower than any other starter with 30+ IP). In particular, Medlen's phenomenal walk rate (3.2% as a starter, 4.4% overall) bodes well. And it's not like this is entirely out of the blue; Medlen dominated the minors with great peripherals, too.

Still, though, we're talking about a pitcher who's 27 and has started 30 games in his MLB career. It's not a lot to go on. Extending Medlen is a high-risk, high-reward scenario for the Braves. The key questions are how much he should get in an extension, and whether the risks outweigh the rewards at that price.

The contract I'm proposing will be a 4-year deal that will buy out Medlen's 3 arbitration seasons (replacing the $2.6 million one-year pact he just signed), plus one year of free agency. I'll also tack on a team option, as is de rigueur in these kinds of deals.

I'll use the same three criteria from the previous posts to determine a fair value for the contract:

• What is the player's present value?
• What are the chances that he can maintain this value in the future?
• How much did comparable players make in their extensions?

Let's start with the simplest of these questions:

What is Medlen's Present Value?

Medlen is scheduled to make $2.6M in 2013 after last week's settlement. Using the 40/60/80 arbitration rule-of-thumb, that implies that his current free-agent value would be around $6.5M per season. Given Medlen's flashes of brilliance, that number seems very low, but it's also pricing in the risks associated with Medlen's inexperience and injury history. If Medlen has another excellent season in 2013 and is still in the arbitration process, he'll get a huge raise as that uncertainty dissipates.

Therein lies one of the Braves' best reasons to extend Medlen now: if he's really a top-of-the-rotation pitcher, he can be locked up now for a fraction of his future price.

Another way of evaluating Medlen's current value is to look at his performance to date. Here are his career stats across 30 starts and 90 relief appearances:

* FIP+, like ERA+, adjusts for park and league and sets 100 as average, with each point above 100 equal to 1% better than average. I converted FanGraphs' FIP- stat to FIP+ for better comparison with ERA+.

With this limited of a sample size, ERA is not as reliable as FIP, so we probably shouldn't assume that Medlen is quite as good as his stellar ERAs suggest. A better baseline for his current value would be that 120 FIP+ as a starter, which implies that Medlen is ~20% better than average. That's plenty good enough, though--right around where Cliff Lee, Johnny Cueto, and Cole Hamels were in 2012.

Since FIP is more reliable than ERA, we're going to trust FanGraphs' version of WAR (which is based on FIP) more than Baseball-Reference's (which is based on runs allowed). There's not much difference between the two systems, however, when it comes to Medlen. FanGraphs says he's been worth 6.5 WAR in his career, including 3.9 last year. B-Ref has him at 6.0 WAR, including 4.3 last year.

Dividing Medlen's WAR figures by 3, we see that Medlen's been worth around 2 WAR per year so far. At $4.5M per win, a 2-WAR pitcher is worth about $9M per season in free agency. If we give Medlen a bit more credit for his stellar rate stats, we might bump that up a bit to around $10-11M per year. Factoring in the arbitration-year salaries, our range is something like $25-30M over the 4 years of the deal.

Next, let's examine the odds of Medlen maintaining or improving on this baseline performance.

What are the chances Medlen will improve his value?

In each of these posts, I've established a set of comparable players that I then used to establish probabilities for future performance. However, Medlen's unique career path results in a dearth of truly comparable players. After much deliberation, I decided to use "breakout" pitchers as my comparison group.

What do I mean by a breakout pitcher? Someone who posted an ERA+ of at least 130 at age 26 (the age Medlen was last year) and who had never come within 20 ERA+ points of his age-26 total before that year. Medlen posted a ludicrous 256 ERA+ last year, compared to his previous best of just 107.

I set minimums of 100 IP and 5 starts for ERA+ considerations and looked at the years 1983 through 2008.* The result was a group of 21 comparables ranging from Greg Maddux to Aaron Heilman. Here's the full list. It's an interesting and diverse group; even if no single player is a good comp for Medlen, the mix should give us a good sense of the various potential outcomes for Medlen's career.

* There were also 6 other players who were too recent to include in the data set: Ubaldo Jimenez, Ricky Romero, Randy Wells, Josh Johnson, J.A. Happ, and Ian Kennedy.

Here are the career averages for the 21 players through their age-26 seasons. I've added Medlen's numbers for comparison. The right column pro-rates the (B-Ref) WAR totals to 200 IP.

While Medlen has been noticeably better than the comp group on a rate basis, he also hasn't pitched as much. This would be true, however, for just about any comparison group that I could come up with. That could mean that any comparison is bound to fail, or it could just mean that we have to assume that Medlen has even higher levels of risk & reward than his comps did.

At any rate, we next need to know how the comps performed in their age-27 through age-30 seasons, the years that will be covered in Medlen's hypothetical extension. Here are the average results (full results at this link):

On average, the group declined slightly in terms of true performance. Their ERA dropped more significantly, but this is just a case of regression to true talent. You can really see based on this group how FIP+ is a better predictor than ERA+ in smallish samples.

While the average player in this group posted 2.3 WAR per "full season" of 200 innings, notice that the average was only about 600 innings over 4 years. In other words, the average player in this group missed an entire season due to either injury or ineffectiveness. This is crucial to our projections for Medlen's future value; in essence, we should only project 3 years worth of healthy production in the 4-year contract.

The average for this group may be somewhat misleading, as the range of outcomes was gigantic and most of the players titled toward one extreme or the other, rather than clustering in the middle as one might expect.

Four of the pitchers in the group continued (or improved on) their breakout years, posting at least 18 WAR in total: Maddux, Dan Haren, Orel Hershiser, and Chuck Finley. (Hershiser was actually a bit worse than he was before, though.) No other player was over 10 WAR, which means that only those 4 were much above average.

Six pitchers were between 5 and 10 WAR, meaning they were roughly average (or a bit below). Another six were between 0 and 4 WAR, which is below average but above replacement level. Finally, five players were below replacement level. The median WAR was only 4.2, while the median ERA+ was 99 (1% below average).

In terms of Medlen's contract extension, it will really only pay off if he produces 10+ WAR, but the comparison suggests that there's only around a ~20% chance of that happening. Then there's another ~30% chance that Medlen earns the contract but little excess value. Finally, there's a slightly more than 50% failure rate (worth 5 WAR or less).

Using either those probabilities or the overall averages, the comparison group implies that Medlen should be paid like a roughly 2-win pitcher for the next 4 seasons. That'd mean a free-agent value of around $9M per season, and around $25M for the 4-year contract.

However, we've already mentioned that Medlen's rate performance thus far is better than most of his comparables; this implies that he has a relatively high ceiling. In other words, we might rate Medlen's chances of reaching stardom at greater than 20%, even if his failure/underperformance chances would remain quite high. If his success rate is 30% instead of 20%, that'd bump his overall projection to something like 2.5 wins per year. That implies a roughly $11M salary in free agency, or around $30M over a 4-year contract.

So far, we seem to be pretty consistently getting a deal in the $25-30M range. How does that compare with other recent pitcher extensions?

How much did players like Medlen get for their extensions?

I looked for pitchers who got extensions at similar points in their careers, and whose deals covered one free-agent season. There have been a bunch of them in recent years, though none of them are great comparables for Medlen, mainly because Medlen is older. I used the extensions signed by Jon Niese, Derek Holland, Gio Gonzalez, Johnny Cueto, Yovani Gallardo, and Ervin Santana: