Much has been said about the recent roulette spins the Cubs find themselves making. The gambles on lower investment, one-year players to be flip at the trade deadline. The international hunt for October players tied to the youthful, high-potential, but high-bust rates found in offshore haunts. The minor league rebuild project that Theo & Jed have stocked directly from the aging carcasses of players once known as Cubs. And now, the final remaining vestige of the 2008 Cubs is one Jeff Samardzija. The front office is prepared for the final bet on black, spinning the Shark, or jumping the Shark, as the case maybe.
So, what is the Shark’s true valuation? Is he promotable as a #1 arm? And given that 2 years of control, what is he worth in trade pieces back? And finally, who should the Cubs target?
Looking at the 2013 season, 56 pitchers were ranked above 2.5 WAR in the MLB, according to Fangraphs. 21 were above 4.0 WAR, for the study, defined as an Ace. #2 starters were 35 remaining arms (4 relievers posted above 2.5 WAR.) Thereafter, 22 pitchers generated a 1.8-2.4 WAR, this is your average #3 pitcher.
The table below shows the averages of various metrics and standard deviations for important pitching categories.
Table. Pitcher Classifications
|2013 Pitcher Type||fWAR(avg)||IP(avg)||xFIP(avg)||K/9(avg)||BB/9(avg)||HR/9(avg)|
|Ace (21 >=4WAR)||5.0||208.5||3.22||8.67||2.30||0.70|
|2013 Pitcher Type||fWAR(dev)||IP(dev)||xFIP(dev)||K/9(dev)||BB/9(dev)||HR/9(dev)|
|Ace (21 >=4WAR)||0.83||17.9||0.41||1.34||0.64||0.20|
|Ace (+1 S.D.)||5.86||226.4||2.81||10.01||1.66||0.50|
|#2 (+1 S.D.)||3.51||211.1||3.30||8.75||1.98||0.62|
|#3 (+1 S.D.)||2.29||191.2||4.30||8.05||3.39||1.22|
|Class of Pitcher||#2||#1||#2||#1||#3||#3|
As you can see, Aces throw more IP, with more K’s, better walk rates, and substantial better HR/rate differential than #2 pitchers. (Standard deviation though for both classes was .2HR/9IP.) None of this is surprising. The bigger question though is: to see group Samardzija actually falls in.
Shark does not rate well on two important factors: HR/9 and BB/9. Meanwhile, the innings and K rates rate as ace-like. What this basically shows he’s got characteristics of 3 levels of pitchers: ace velocity/stuff (K rate) and ability to throw innings; a low #3 starter on walking (-1 S.D.) and #3 level for HR rate (1.05 is below average for a #3). His xFIP is good enough to be in the class of a #2.
So, Shark is, at best, a solid #2 starter. He is projected as a 3.1 WAR using Fangraphs Steamer, this is contrary to prior statistics as he seems incapable of reducing his home run rate or his walk rate down to ace-like levels. His BABIP (.314) hurt him slightly (.290 for the league) in 2013 while his GB rate (48.2%) does bode well. Shark’s 71.7 LOB% is below what aces typically average 76.0+/-3.5%, but more like #2-3 starters.
Valuation of Shark
And so what is a #2 starter now worth to a team in the hunt?
If Samardzija is traded before the season starts, his reasonably expected contribution for two years should be at 5.5-6.0 WAR (65%), with an outside probability of 8.0WAR(20%) and 2.5 WAR(15%) if an injury should take place, like Tommy John, losing a year. This produces a weighted average (risk adjusted) of 5.7WAR.
The cost to receive this rate will be around 35% of the free agent price level in year 5, and 50% of free agent price in year 6, if we set WAR at $6.1M/WAR. As Shark has a short track record of success, his WAR base will be 2.5 and 3.5 for those seasons. Recently, Samardzija was valued at $5.35M – my rough 2014 calculation at 35%2.5(eWAR)6.1M equals 5.338M. His 2015 value at 3.5WAR would be roughly $10.675M, which lines up to the high side with Jeff Sullivan’s recent article that mentions Shark’s 2015 arbitration number. (These too are just working projections and would take a different model to work up right.)
From there, you can look at the yearly calculations run based on probable performance scenarios. Ace Jeff was set at 20% outcome. This is fair, if you think he’s capable of such a move up. Normal Jeff is at 65%. This is due to his steady enough nature velocity wise and the lack of overall hard miles on the arm. Injured Jeff was put at 15%. Baseball Prospectus did a study on injury factors. Jeff, never being hurt, would be about a 4.9% likelihood for injury, but for the fact he threw 3,425 pitches in 2013 and according to Pitcher Injury Factors, had some slight spikes in errant release point late in the season. This increases his likelihood of injury. So I padded the number to 15% based on this paragraph:
To give you some estimate of the effect that might have, imagine that a pitcher went from 3000 pitches in a season to 3300 (the equivalent of going from 30 starts with 100 pitches per start to 110 pitches per start). The increased chance of a DL visit is on the order of a couple of percentage points. Given that the baseline rate for a pitcher who is not previously injured is 4.9 percent, that’s not trivial. Managers, please see to it that your pitchers never throw another pitch.
|$/WAR on FA Market||$ 6,100,000|
|Extension @2.5 WAR||$ 15,250,000|
|Arb Shark Salary||$5,337,500||$10,675,000||$16,012,500|
|WAR by Pitcher Type||Prob||WAR Production||Total|
|Shark’s WAR Ace||0.2||4||4||3.5||3||2.5||17|
|Shark’s (WAR) Norm||0.65||2.875||2.875||2.5||2||1.5||11.75|
|Shark’s WAR Hurt||0.15||1.5||1||2.5||2||1.5||8.5|
|Arb (Surplus Val)||Extended||AAV of Contract|
|Ace Surplus Val||$32,787,500||Ace Shark||$4,485,294|
|Norm Surplus||$19,062,500||Normal Shark||$6,489,362|
|Hurt Surplus||-$762,500||Hurt Shark||$8,970,588|
If for 2 seasons, Shark produced at “ace-like” levels of 4.0 WAR or above, his surplus value to a team is $32.8 M. This is what would be needed (above his $16.1M arbitration salary) to acquire WAR in the Free Agent market. Even “normal Jeff” has a $19.0M dollar incentive for the acquirer. This is what we are looking to get from a prospect back – value in the form of $19-33M of production, risk adjusted value (as to the probability such top prospects don’t work out – we need more to satisfy that surplus value correctly.)
If a team wants to extend him for say $15.25M for 5 seasons and he produced his ace level over that time frame, it would be well under-market rate at $4.485M. Normal Jeff though is slightly overpaid – but WAR values for free agents will rise with baseball’s inflation, making this almost a moot point. Hurt Jeff will hurt anyone.
This is the calculation the Cubs have been doing: to get him in the $60M range (expecting NORMAL Jeff), while marketing him as a 3.5+WAR per season upper tier #2 pitcher, Ace Jeff.
The Golden Tier of Prospects: Derived from a Draft Model
So what prospects could the Cubs receive?
The very best prospects have potential to generate substantially better WAR than Samardzija has to date, or will in the future. However, to identify said winners is no easy task.
I started with a logical assumption (to me): that the value of top prospects pretty much follows the same as value curve as picking the 1st year draft, ONLY amplified to reflect how much further along these players are in the process of becoming big leaguers. I looked at how to adjust this out to a level that reflects the nature of said progression. I started by selection two ideas: one was the top WAR producer for the years 1990-95 in the 1st round only of those drafts. This set the peak value (averaging those top numbers out). Call that a start point – a HOF player (74.83 WAR) is currently in the minor league top prospect bin. I almost guarantee it. The goal is to find this gem.
The second, was to find a curve that models the decline from this hidden HOF star down to the 300th best prospect in the minors, a guy who is in the top 6.7% of the entire minor league pool of (say 4,500?), but isn’t that special compared to 750 big leaguers (he could have 4-1WAR years). I found that with this 2009 study by Sky Andrecheck, titled, “Draft Picks and Expected Wins Above Replacement“. To me, it just reflects the pyramid of the talent pool getting smaller, while the expected value increases. And it makes some sense: about 10-15 guys will make substantive impacts per season in the majors (1 per two teams) – while maybe another 40 will get to stay on due to natural turnover of talent. Older players leaving, and injuries. Tom Tango commented on the analysis, so it can’t be that off.
Sky’s model baseline formula was this:
WAR= a * (selection#)^ b, where a and b are the parameters of the model.
Running a non-linear regression, we find those parameters equal to a=19.8 and b=-.50. The model fits very well as you can see from the graph [below].
Replace selection # with prospect ranking, and the graph appears like this:
My Top prospect shift (to the right) garnered this graph, which is truncated some for easier insertion.
I feel the Cubs have secure about 10WAR to break even on this trade. That which is lost (Samardzija at 5.71 expected WAR) + a 4.25 WAR premium based on the most likely positive scenarios of value surplus at $25.9M average or 4.25 WAR. (15% injury risk premium – he gets hurt – is born by the acquirer. It is a sport.)
So to get 10WAR, you could just say a mid-50s prospect with a 100% success rate will accomplish it, but that is implausible; and also ignores the slim chance he does that in just 6 control years.
No, a scenario that likely gives you what you want has to be the success rate for a prospect class times the expected average WAR for that group. Something like this – although, the success rate is something to discuss.
|Tier||Prospects||Avg. WAR*Success Rate||Success Rate|
In short, the Cubs could feasible ask for four 15-60 mid-tier prospects, if a team actually had that number and wanted to part ways. (No one does – so that means you have to barter some.) But the likely reason no one wants to give up a top tier prospect for Samardzija is that it is an OVERPAY. Two prospects is really hard to swallow – a top 15 plus a top 60 would be a 5WAR differential to the Cubs. Yet, if someone is desperate, they may overpay later at mid-season, assuming Shark is healthy and doing well. So too, the injury risk premium shifts back to the Cubs….
It takes two to tango, or gamble, as the Cubs will do.