---

Wednesday, December 8, 2010

What Cost Cliff Lee?

The rumor mill is continuing to swirl around the Yankees’ number one target this offseason.  It’s tough to know what’s really true and what’s been strategically planted by representatives of all the interested parties, but that doesn’t mean we can’t try and objectively determine how much Cliff Lee should be paid.
The most important considerations for that should be

a) How good is Cliff Lee and how good will he be going forward?

b) How much is that worth to the team that wants to sign him?

To answer the first question, I’ll try and use my CAIRO projection for Lee, but you can use whatever system you want.
Here are my range of projections for Lee in 2011 within one standard deviation of his baseline performance.  Like any pitcher, there’s the risk of an injury making him worth nothing, so although I won’t really get into that here it should be at least a slight consideration.

%	G	GS	IP	H	HR	BB	K	RA	ERA	FIP	RSAR	WAR
80%	32	32	220	204	16	31	190	3.53	3.22	2.87	60	6.0
65%	30	30	210	201	17	32	175	3.72	3.41	3.10	52	5.2
Baseline	29	29	200	198	18	33	161	3.91	3.59	3.34	46	4.6
35%	26	26	180	184	18	32	140	4.11	3.77	3.58	37	3.7
20%	23	23	160	168	18	31	120	4.30	3.96	3.81	30	3.0

FIP: Fielding-independent pitching
RSAR: Runs saved above replacement level (using RA)
WAR: Wins above replacement level (RSAR divided by 10)

Lee’s a very good pitcher, and he projects to be one again in 2011.  But I don’t think he’s the best pitcher in baseball, even though some in the media talk about him like he is.  Whatever the case, there’s little doubt that he’d help any team in 2011, except for a team that already has say, 13 or so aces.  Actually, make that 12 now.

The problem is, you can’t sign Lee for one year.  Well, I guess you could if you offered him $50M or something, but that’s not particularly smart.  So then we need to try and think about how much he’ll be worth over however long you have to sign him for.

A generally good rule of thumb as researched by Tango Tiger is to assume a player in their 30s will decline by about 0.5 WAR per season.  This is a general aging decline that includes some risk of injury/attritition as well as the possibility that some players will play better than expected.  Of course, any individual player will not follow the general aging rules, and it’s not necessarily a nice linear progression, but if you are trying to forecast out four or five years it’ll probably be close more often than not.

Here are three scenarios with Lee.

Scenario 1: Lee is about as good as his baseline projection thinks he is.

I actually think Lee’s probably better than his baseline CAIRO projection.  Not 7.0 WAR good mind you, but maybe 5.0 WAR good.  But since it’s the baseline, it’s the first scenario.

Year	Age	IP	H	HR	BB	K	RA	ERA	FIP	RSAR	WAR	tWAR
2011	33	200	198	18	33	161	3.91	3.59	3.34	46	4.6	4.6
2012	34	190	193	18	32	148	4.07	3.73	3.41	40	4.0	8.6
2013	35	181	188	17	30	137	4.21	3.86	3.47	35	3.5	12.1
2014	36	172	184	17	29	124	4.32	3.96	3.60	31	3.1	15.2
2015	37	163	180	17	28	115	4.48	4.11	3.68	27	2.7	17.9
2016	38	155	176	16	27	108	4.69	4.30	3.74	22	2.2	20.1
2017	39	147	172	16	25	98	5.01	4.59	3.85	16	1.6	21.7

tWAR: Total WAR of contract through year n

Assuming Lee can pitch seven more years may be optimistic, but since there are at least rumors out there that some teams are considering offering him a contract of that length here’s what it may look like.  I realize there’s a big gap here between FIP and ERA as we move on, but from what I’ve found BABIP skill does decline as a pitcher ages.  The main thing we need to focus on is wins anyway, so the shape of how his value accrues is a secondary concern.

A win is worth a certain amount to each team.  For a team that’s on the playoff bubble a win is worth more than for a team that’s not.  As the largest revenue team in baseball, a win is worth more for the Yankees than any other team, which is why they can usually outbid anyone for a player they really want. 

I have no idea how much a win is worth to the Yankees right now.  We can try and estimate based on what they’ve paid free agents (say Jeter and Mo just now), but that means we need to know how much they value them and frankly both of them are almost certainly being paid more than they’re worth due to their iconic status.

So here’s a table that shows how much Lee would be worth for the Yankees based on four different valuations of a marginal win.

$/WAR		$4,500,000	$5,000,000	$5,500,000	$6,000,000
Year	tWAR	tValue	tValue	tValue	tValue
2011	4.6	$20,552,226	$22,835,807	$25,119,388	$27,402,968
2012	8.6	$38,586,878	$42,874,308	$47,161,739	$51,449,170
2013	12.1	$54,447,865	$60,497,627	$66,547,390	$72,597,153
2014	15.2	$68,599,324	$76,221,471	$83,843,618	$91,465,765
2015	17.9	$80,749,697	$89,721,885	$98,694,074	$107,666,262
2016	20.1	$90,671,401	$100,746,001	$110,820,601	$120,895,201
2017	21.7	$97,750,955	$108,612,172	$119,473,389	$130,334,606

$/WAR Estimated value of one win above replacement (WAR)
tWAR: Total WAR of contract through year n
tValue Total Value (tWAR times $/War)

If this is accurate, Lee will almost certainly be overpaid by any team who values a marginal win at $6M or less.  On average, teams spend $4.5M per marginal win on the free agent market, so that probably covers just about every team.

Scenario 2: Lee is about as good as his 80% projection thinks he is.

Year	Age	IP	H	HR	BB	K	RA	ERA	FIP	RSAR	WAR
2011	33	220	204	16	31	190	3.53	3.22	2.87	60	6.0
2012	34	209	198	15	29	174	3.67	3.35	2.93	53	5.3
2013	35	199	193	15	28	162	3.80	3.47	3.00	48	4.8
2014	36	189	189	15	27	146	3.90	3.56	3.12	43	4.3
2015	37	179	185	14	26	135	4.04	3.69	3.19	38	3.8
2016	38	170	181	14	25	128	4.23	3.86	3.24	33	3.3
2017	39	162	177	14	24	116	4.52	4.12	3.35	26	2.6

$/WAR		$4,500,000	$5,000,000	$5,500,000	$6,000,000
Year	tWAR	tValue	tValue	tValue	tValue
2011	6.0	$26,805,561	$29,783,957	$32,762,352	$35,740,748
2012	11.3	$50,791,595	$56,435,106	$62,078,616	$67,722,127
2013	16.1	$72,315,542	$80,350,602	$88,385,662	$96,420,722
2014	20.4	$91,853,404	$102,059,337	$112,265,271	$122,471,205
2015	24.3	$109,130,160	$121,255,734	$133,381,307	$145,506,880
2016	27.5	$123,933,587	$137,703,985	$151,474,384	$165,244,782
2017	30.1	$135,667,648	$150,741,831	$165,816,014	$180,890,197

I find this scenario to be somewhat unrealistic, but I’m presenting it anyway.

Scenario 3: Lee’s back problems or some other decline leading to him being only as good as his 20% projection.

Year	Age	IP	H	HR	BB	K	RA	ERA	FIP	RSAR	WAR	tWAR
2011	33	160	168	18	31	120	4.30	3.96	3.81	30	3.0	3.0
2012	34	152	164	17	29	110	4.47	4.11	3.88	25	2.5	5.5
2013	35	144	160	17	28	102	4.62	4.26	3.95	22	2.2	7.7
2014	36	137	156	16	27	92	4.74	4.36	4.08	19	1.9	9.5
2015	37	130	153	16	26	85	4.91	4.52	4.17	15	1.5	11.1
2016	38	124	150	16	25	81	5.14	4.74	4.24	11	1.1	12.2
2017	39	118	146	15	24	73	5.49	5.06	4.35	6	0.6	12.8

$/WAR		$4,500,000	$5,000,000	$5,500,000	$6,000,000
Year	tWAR	tValue	tValue	tValue	tValue
2011	3.0	$13,388,608	$14,876,232	$16,363,855	$17,851,478
2012	5.5	$24,799,662	$27,555,180	$30,310,698	$33,066,216
2013	7.7	$34,523,462	$38,359,402	$42,195,342	$46,031,283
2014	9.5	$42,956,443	$47,729,381	$52,502,319	$57,275,258
2015	11.1	$49,832,126	$55,369,028	$60,905,931	$66,442,834
2016	12.2	$54,940,724	$61,045,248	$67,149,773	$73,254,298
2017	12.8	$57,734,147	$64,149,053	$70,563,958	$76,978,863

This is the real concern.  The fact is, aside from his Cy Young season in 2008 Lee’s never been the clear-cut best pitcher in the league, unless you want to assume his FIP tells us everything we need to know about him, a position I refuse to accept.  He appears to be in very good shape, but he will be 33 and he has had back problems and those are legitimate and real concerns.

For that reason, I’d rather see the Yankees overpay for five seasons and avoid the very real risk that they will be paying for a Lee that is barely above replacement level in years six and seven if scenario 3 comes close to passing.  But given the way things are shaping, he’s probably getting six years from someone.  If that’s the Yankees, it’s almost certainly good news for 2011, but could become a real problem after that.