Long-term projections of Royals prospects: Part one

It seems like there is a lot of work being done right now about projecting prospect and evaluating their future worth. Whether that worth is expressed through wins above replacement or a dollar value, there's always going to be an interest towards the games (hopeful) future players.

Groundwork has been laid through multiple articles covering anything from the cost of a draft pick, to KATOH projections, to prospect failure rates, to assigning a dollar value for a prospect ranking. One thing I like looking forward to every year around this time is the PECOTA  long-term forecast of players. While obviously it isn't an exact science in it's results, projections never are, it's one of the few projection systems that projects players (not just prospects) on a further than one or two year basis. For instance, the Oliver projection systems goes out to five years, but in PECOTA's case, it goes ten years.

So we'll look at some of the Royals better (relative to at least other Royals prospects), prospects. I've spilled a lot of digital ink on these pages about subscribing to Baseball Prospectus (I do not work there obviously), and I'll implore you one more time to do so. I'll try to reason some of the results as best I can. I'm not a PECOTA expert and it's seen many tweaks and iterations over the years and things might be changing that we don't know about.

Also, BP has a stat called True Average which I'll include in the tables. True Average, as described by BP:

True Average incorporates aspects that other linear weights-based metrics ignore. Reaching base on an error and situational hitting are included; meanwhile, strikeouts and bunts are treated as slightly more and less damaging outs than normal. The baseline for an average player is not meant to portray what a typical player has done, but rather what a typical player would do if given similar opportunities. That means adjustments made for parks and league quality. True Average's adjustments go beyond applying a blanket modifier-players who play more home games than road games will see that reflected in their adjustments. Unlike its predecessor, Equivalent Average, True Average does not consider baserunning or basestealing.

Here is an example of the True Average spectrum based upon the 2009-2011 seasons:

Excellent - Miguel Cabrera .342
Great - Alex Rodriguez .300
Average - Austin Jackson .260
Poor - Ronny Cedeno .228
Horrendous - Brandon Wood .192

One must also note that these projections do not include 2015 projections, and start immediately for 2016.

Raul Mondesi:

Year	Age	PA	HR	SB	TAv	WARP
2016	20	250	3	6	0.214	-0.9
2017	21	412	6	11	0.221	-0.2
2018	22	593	9	16	0.234	0.9
2019	23	554	8	14	0.228	0.4
2020	24	562	9	15	0.231	0.6
2021	25	553	9	15	0.231	0.6
2022	26	573	9	15	0.236	0.9
2023	27	576	11	15	0.237	0.9
2024	28	580	10	14	0.238	1
Sum/Average		4653	74	121	0.230	4.2

So that's not exactly what you want to see out of the Royals arguably best prospect, but there are some things to note. First, Mondesi hasn't played above A+. Generally stats don't have a high correlation to major league success until AA/AAA. Also Mondesi is an extremely young player for his level. He'll be 19 years old in the Texas League in 2015. The last 19 year old to get a qualified amount of PA's in the Texas League was higher lauded shortstop prospect Jurickson Profar in 2012. There have only been 3 such cases since 2006 (Profar, Mike Trout, and Elvis Andrus).

One other caveat here is that PECOTA has Mondesi as an average defender (0.0 FRAA). Most scouts peg Mondesi as being an above average defensively shortstop and perhaps one of the better defenders in all the minor leagues. PECOTA doesn't inherently know that and that skews the projections as it labels him as an average defender at SS.

Here's the possible differences there:

Year	Player	TAv	wRC+	Def	WAR
2011	Yuniesky Betancourt	0.230	72	-0.1	0.0
2011	Alex Gonzalez	0.231	75	6.6	0.8
2012	Yunel Escobar	0.230	73	10.8	1.4
2012	Alexei Ramirez	0.229	72	15.4	1.9
2014	Andelton Simmons	0.230	71	22.1	2.3

So we have multiple levels of defensive value with basically the same TAv/wRC+.  For about every 5 runs above average defensively a player is, that's worth about 0.5 fWAR (in the above graph).

So where do you put Mondesi? If you think he's a 10+ run defender then you could potentially boost his WARP by a full win or so. Is he a maybe 15-20 run defender? That might be pushing it a bit maybe on an annual basis at least. Mondesi is a very good defender, but defensive metrics can be inconsistent and fickle at times. Simmons, arguably the best defender in baseball, has been worth anywhere from 13 to 31 runs above average the past three years.

Comparable player: Alcides Escobar

Escobar is a regarded defender by scouting and but the statistics have been hit and miss on him and volatile. My focus is mainly on the bat here though as he has a career .235 TAv (similar to Mondesi's projection) and is a low avg and OBP guy with a little blip of power at SS.

Sean Manaea

Year	Age	IP	K/9	BB/9	ERA	WARP
2016	24	187	8.2	3.6	4.24	0.3
2017	25	161	9.0	3.1	3.60	1.7
2018	26	157	8.9	3.1	3.81	1.2
2019	27	157	9.0	3.0	3.39	2.1
2020	28	158	8.9	2.9	3.42	2.1
2021	29	151	8.6	2.9	3.58	1.6
2022	30	152	8.8	2.7	3.45	1.9
2023	31	155	9.0	2.6	3.39	2.1
2024	32	152	8.9	2.4	3.27	2.3
Sum/Average		1430	8.8	2.9	3.57	15.3

One must be aware that WARP (Wins Above Replacement Player) is different than FanGraphs fWAR for pitchers in that WARP uses earned runs while fWAR uses FIP. I also believe their replacement level is higher as a full roster of 0 WARP players would win a little over 50 games (fWAR/rWAR uses 47.7 games), so 2 wins in WARP is slightly above 2 wins in fWAR.

You can debate the philosophies and criteria of calculating pitcher value, but 15 wins over 10 seasons is 1.5 wins per season. That's a below average player.

Comparable player: TBD

I prefer using FIP rather than ERA to estimate pitcher value so let's try to convert it to some similarity of a player using FIP based wins.

If we've got an 8.8 K/9, 2.9 BB/9, and a 0.70 HR/9 (the career averages of Manaea in the above projections) then here are some similar players (stats are from 2010-2014).

Name	IP	K/9	BB/9	HR/9	ERA	FIP	WAR
Jon Lester	1038	8.41	2.96	0.82	3.54	3.48	22.4
Yovani Gallardo	969.1	8.37	3.11	0.97	3.73	3.68	13.7
Ian Kennedy	1006.2	8.25	2.89	1.04	3.81	3.85	13.6
Josh Johnson	516.2	8.54	2.84	0.66	3.4	3.15	11.8
Lance Lynn	616	8.75	3.26	0.67	3.46	3.34	9.8
Jeff Samardzija	715.1	8.71	3.11	0.93	3.74	3.65	9.7
Tommy Hanson	580.1	8.25	3.15	1.05	4	3.94	6.9
Sean Manaea	1430	8.80	2.91	0.70	3.57

Lester is a fellow lefty like Manaea, and he struggled with his command in his first few years but still had excellent results, including a 5/6 win season by age 24/25. That's probably not going to happen for Manaea (he'll be 23 this year - Lester was a HS draftee), but they share similar career stats vs Manaea's projections.

Name	IP	K/9	BB/9	HR/9	ERA
Jon Lester	1038	8.41	2.96	0.82	3.54
Sean Manaea	1430	8.80	2.91	0.70	3.57

Samardzija was a college pick who was raw (he played TE at Notre Dame) and took some time to sculpt. He was very inconsistent and poor pitcher until he took a gigantic step forward at age 27, something that PECOTA thinks Manaea may do (though not as large as Shark's).

None of the players above share a pitch repertoire as Manaea (who's FB/SL/CH). Lester throws a curveball as his fourth pitch while Samardzija doesn't throw a changeup.

Kyle Zimmer

Year	Age	IP	K/9	BB/9	ERA	WARP
2016	24	190	8.3	3.1	4.12	0.6
2017	25	130	7.7	3.2	3.34	1.8
2018	26	164	8.9	2.9	4.06	0.7
2019	27	133	8.4	2.7	3.93	0.8
2020	28	116	8.2	2.9	4.06	0.5
2021	29	115	8.3	2.7	3.82	0.9
2022	30	112	8.3	2.5	3.81	0.9
2023	31	106	8.3	2.5	3.81	0.8
2024	32	99	8.1	2.5	3.79	0.8
Sum/Average		1165	8.28	2.78	3.86	7.8

PECOTA does know about injuries, especially for pitchers, and has been known for it's injury-cautious shell. BP has a pretty extensive record keeping of injuries, even for prospects, and surely PECOTA sees a low amount of innings in 2014 and Zimmer's actual injury. Does PECOTA care about if Zimmer didn't pitch because of say suspension rather than injury? I don't know.

Comparable player: Ian Kennedy

That's not a very flashy comp, but Kennedy has been a good starter for most of his career. He also dealt with a shoulder injury in 2009 (at age 24) that wiped away a year before returning pretty strong in 2010. At one point Kennedy shared the same rep as Zimmer with a FB/SL/CB/CH before ditching the slider in exchange of a cutter (something Zimmer isn't going to do).

Player	IP	K/9	BB/9	ERA	fWAR/WARP
Kennedy	1006	8.25	2.89	3.81	13.6
Zimmer	1165	8.28	2.78	3.86	7.8

Miguel Almonte

Age	IP	K/9	BB/9	ERA	WARP
23	171	6.5	3.3	4.12	-0.3
24	140	6.5	3.0	3.34	0.3
25	118	6.4	2.8	4.06	0.1
26	134	7.0	2.7	3.93	0.3
27	111	6.6	2.8	4.06	0.4
28	106	6.5	2.9	3.82	0.2
29	111	6.8	2.4	3.81	0.6
30	109	6.8	2.2	3.81	0.7
31	103	6.9	2.3	3.79	0.7
Sum/Average	1103	6.67	2.71	3.86	3

I'm low on Almonte myself (had him as our 7th best prospect) but he has a useful changeup and fastball,and even if the curveball is inconsistent he's got the ability to be in the rotation. Ultimately the curveball may never stay average, rather than flash as it does now, and he could move to the bullpen where his 93-94 MPH fastball could play up slightly.

Comparable player: Jeremy Hellickson

Hellickson was always one of those guys who's ERA outperformed his FIP and we all wondered when it would catch up with him. Well after winning Rookie of the Year in 2011, it finally did in 2013 and now his FIP is much better than his ERA.

Player	IP	K/9	BB/9	ERA	fWAR/WARP
Hellickson	640	6.51	2.95	3.78	4.7
Almonte	1103	6.67	2.71	3.86	3

This is more of a statistical comparison than actual stuff comparison (like most others in this article). Hellickson features a low-90's fastball (average of 90.8 MPH) while Almonte is more 93-94 MPH. Like Almonte though, Hellickson use his changeup often and has been one of the better changeups in baseball.

You'll always have to remember that's there going to be a disconnect between scouting reports and projection systems as well as possibly in the stats/production itself (which further changes projections). The next few prospects will be in another post coming soon.