moonslav59

Good point. I remembered incorrectly and felt he caught much more than just 20 innings more in 2014 than his scattered 2016 innings. 20 innings more is not a significant sample size difference. My bad.

Stop saying "It won't mean anything." Nobody is saying when player sucks in the playoffs, it is meaningless. While post season sample sizes are rather small, it does not mean they meaningless. % post seasons are smaller than 10, so 10 sucky post seasons would hold more "meaning" than just 5, and I suppose you could begin to judge the guy as a "choke" after a certain amount of games played, but the poor playoff performance might have little to do with choking under pressure. It could just be bad luck. It could be lack of stamina after 200 or 220 IP'd during a season. It could just be nothing but a selected choice of his worst games that happened to fall during the playoffs. One could probably find a sample size throughout Prices career where he has an 8.50 ERA between 7pm on April 22nd and 4 pm on April 24th. Would anybody care? Would you be concerned with his next start on April 23rd? I realize I am oversimplifying things, and I'm not denying that a player can "choke" under pressure or "rise to the occasion" as well, but there is a point where the sample size is too small to know it's just coincidence or if there is a deeper meaning or reason involved.I'm not sure if Price's playoff sample size is large enough to label him anything. I, for one, would not bet on him in his next playoff start, so I wouldn't say his history has no meaning, but what if he pitched two straight shut outs next October, what would that mean to you guys? He's no longer a choke? He's still a choke, because his overall numbers still suck. Would his two game sample size now be large enough to expect continued brilliance? If forced to bet, what sample size do you use to determine how well he'll do in his next game?

Would you really say a player going oh for 1 "sucked"? Even if his out was a long fly ball robbed of an HR by a spectacular play? How about oh for 2? We probably all have differing views on when a sample size become "definitive" or more definitive. Does going of for 2 really define a player? O for any 25 stretch in his career? Maybe a 10 for 100 stretch for a good hitter might define him as "streaky", but I don't think a 2 for 20 stretch necessarily does, unless he has a lot of them over his career.

None of us have ever said it "doesn't mean anything". Any 0-25 streak has meaning and can greatly affect a team's chances of winning. When I say "making definitive judgments" I mean labeling a player as a choke based on just a 25 AB sample size. That's "definitive" to me. Saying he sucked for those 25 ABs is not "definitive" in the sense that we are making a long-lasting judgment on a player's overall skill level, career accomplishments or expected future production level. Maybe we are at odds with the semantics or contexts associated with the words "definitive" and "judgment". Perhaps, I should have been more descriptive in my explanation, but then I'd probably be accused of being too verbose.

Agreed, but is someone who goes 1 for 5 in a WS as bad as guy who went 10 for 50? Yes, you can say both sucked, but a sample size context is needed in both examples. Vaz had pretty good pitch framing skills in 2014... a much larger sample size than 2016. It's hard to know why he was worse in 2016. maybe it was partially due to injury, not fully being rehabbed and/or catching a staff of pitchers he had little experience with. Yes, we can say he was "bad" in 2016, but can we really say for certainty that because of his smaller 2016 sample size, he's a bad pitch framer? Or, more importantly, how can we view the likelihood of his being a bad pitch-framer going forward. My guess is that with added experience with newer pitchers on the staff, a healthier body, and hopefully improved confidence, he will revert back to be a plus pitch-framer, a plus CS% catcher, a plus wild pitch-blocker, and a better overall CERA-related catcher than he was in 2016. His hitting is still a mystery, but I'm not optimistic he'll ever be consistently over .725 or even .700.

I totally get your point and don't disagree. Think of it this way: Say a very consistent .300 hitter (hit between .295 and .305 for 4 straight years and is 29 years old) is going through a 5 for 50 slump (.100). Over his next 10 ABs would yo bet his BA would be closer to .100 or .300? How about over his next 50 ABs? His next 500 ABs? If forced to bet, I'd bet closer to .300 on all three, but I'd feel much safer the larger the sample size going forward. I'm not a betting man (anymore), but I'd bet this hitter hits over .200 in his next 500 PAs, unless his name is Allen Craig.

The sample sizes are really too small to make any definitive judgment on CV's 2016 defense. On CERA, almost every pitcher has vastly different sample sizes with any specific Sox catcher. I provided the CERA numbers pitcher-by-pitcher below, but please note some of the vert tiny sample sizes. Here are a few things I noticed: Out of our top 6 IP pitchers (starters for the most part): Leon had a better CERA than Vaz with 3 (Price, Porcello & Buch). Vaz had a better CERA than Leon with 2 (Wright and ERod). It may not be a fluke that Vaz had caught both of these guys in the minors. When looking at the other 5 pitchers (RP'ers) with over 48 IP, we see a much different result. (Note: the sample sizes are even tinier here.) Leon had a better CERA with only 1 RP'er (Ross with an IP difference of 27 to 7) Vaz had a better CERA with the 4 others (Barnes, Kimbrel, Hembree & Tazawa- all with both pitchers having a sample size of 15 to 25 IP) (Note: Leon did better with Uehara and Kelly, but his IP with Kelly was just in 7 IP, and neither had over 18 IP with Uehara.) If we use these numbers to allign the best catcher with our starters next year, we might want to use Vaz as Wright's "caddy" and possibly with ERod and/or Pom. Here are the 2016 CERA numbers I compiled: L=Leon V=Vazquez Ha=Hanigan Ho=Holladay S=Swihart (I bolded the leader(s), if the sample size is large enough to be top 2) Price: L 3.23-109/ V 4.62-97/ Ho 4.15-13/ S 5.73-11 Porc: L 2.52-107/ V 3.64-94/ Ha 3.45-16/ S 6.00-6 DWri: L 6.46-24/ V 2.72-36/ Ha 2.90-90/ S 1.35-7 Buch: L 3.84-61/ V 5.60-53/ Ha 3.24-8/Ho 2.25-8/S 10.00-9 ERod: L 4.90-79/ V 4.50-22/Ha 3.00-6 DPom: L 5.85-20/ V 0-0/ Ha 7.88-8/ Ho 3.32-41 (Norris 1.56-81/Bet 6.00-21) Barnes: 6.04-25/ V 1.37-20/ Ha 5.17-16/ Ho 0.00-5/ S 3.38-3 Ross: L 1.65-27/ V 12.86-7/ Ha 1.98-14/ Ho 0.00-3/S 5.79-5 Kimb: L 3.20-20/ V 2.04-18/Ha 1.74-10/ Ho 10.80-3/ S 13.50-2 Hemb: L 3.18-17/ V 2.08-17/Ha 1.98-14/Ho 3.00-3 Taza: L 4.24-17/ V 4.11-15/ Ha 1.38-13/ Ho 18.00-2/ S 3.86-2 Ueh: L 3.44-18/ V 5.94-17/ Ha 0.00-8/Ho 0.00-1/ S 0.00-3 Kelly: L 1.35-7/ V 4.50-10/ Ha 6.97-21/ Ho 3.38-3

Interesting use of the word "floating". I remember feeling the same way in a championship game my hall won while playing in an inter-mural league at the University of Notre Dame. I too was usually an assist and defensive guy, but that day I was "floating" in a "zone". I scored 14 of my team's 25 baskets. I never thought about what was going on at the time, and as I read later, that's what probably helped me "stay in the zone." I remember when it was my turn to be rotated out for a sub, the sub said to me, "Are you kidding me? I'm not going in for you!" Later, I read a book called In the Zone: Transcendent Experience in Sports. I highly recommend it to any true sports fan. It based on a study of great players (mostly in golf and tennis) and what they said after having a fantastic performance. The language used all sounded very similar. "I was in a zone." "I was floating." "It was like I was outside my body." "Everything was like it was in slow motion, and I could see ahead." "I wasn't thinking about anything." "As soon as I became aware of how well I was doing, the thought distracted me, and I left the zone." I've lived and seen "hot" and "cold", and although there may not seem to be any or much randomness to it from the player's point of view, I think some or a lot of "randomness" comes into play as to when the point of entering or leaving "the zone" occurs. It could be something as simple as a mere thought in the mind that changes the focus. It could be a minor injury or a personal issue. As a fan, trying to determine when a streak begins or ends is, in a way, random. I'm not trying top ut words in anyone's mouths, but that is what I think people meant by "random". Some players are known for being very streaky- good and bad, like Mike Napoli. Mike had super seasons and average seasons and a couple bad ones too. Within a season, he had up and down stretches that seemed endless. I remember arguing with several [poster who wanted him benched in 2013 during a long "slump". I kept saying, we may be taking him out just as his hot streak was about to begin. I said that for a while, as his slump continued (longer than I expected), but he did finally bust out and helped us win a ring. Maybe with a player like Mike, it's safer to bet on him continuing a streak (hot or cold) than to bet on him to "return to norm", since he hits at his norm for extended periods of times at a much lower rate than other players. Other players are much more consistent. They have shorter slumps and usually shorter hot streaks as well. They are more "predictable" or "random" in the sense that they hover around their career norm pretty consistently in almost any selected shorter sample size within their career. I would not bet on a guy like Pedey to continue in an ongoing slump. As far as I know, there are no studies that show that a baseball player's most recent 20, 30 or 50 game results are a better predictor of what will come in the next 20,30, 50 or 162 games than using the players last 1-3 year sample size or career norm as a predictor. In fact, I'd bet the opposite is true. Using some sort of combination between recent 1 year, 3 year and extended trends of legitimate sample sizes is probably the best predictor of what to expect going forward.

Owens or Johnson. It's about time one of these guys stepped up their game. Both had close to a 50-50 chance of becoming at least a decent #5 starter at some point in their career. That gave us a 75% chance that at least one could be "that guy". We've seen squat so far. It's time to prove your worth guys! (I'd like to see us sign a couple vets to minor league deals as added depth.

Link?

Great article on the value of pitch framing: http://www.fangraphs.com/blogs/how-bad-could-a-pitch-framer-possibly-be/ Another view: http://www.hardballtimes.com/pitch-framing-was-doomed-from-the-start/

Here's how I see our positional depth chart right now: C Leon, Vazquez, Swihart (AAA), Butler (AAA) 1B Ramirez (DH)- Moreland, Travis (AAA), Holt, Craig (AAA non 40 man) 2B Pedroia, Holt, Hernandez (AAA), Rutledge, Marrero (AAA) 3B Sandoval, Rutledge, Holt, Hernandez (AAA), Domingues (AAA) SS Bogaerts, Hernandez (AAA), Holt, Marrero (AAA) LF Benintendi, Young, Holt, Moreland, Swihart (AAA) CF Bradley, Benintendi, Betts, Young, Holt, Castillo (AAA non 40 man) RF Betts, Bradley, Young, Holt, Moreland (assuming Brentz is not on the 40 man roster by opening day) DH HanRam-Young, Swihart (AAA), Witte (AAA non 40 man) SP Sale, Porcello, Price, Rodriguez, Pomeranz, Wright, Kelly, AAA: Elias, Johnson & Owens RP Kimbrel, C Smith (DL), Thornburg, Wright, Kelly, Ross, Hembree, Abad (DFA/DL?), AAA: Scott, Elias, Workman, Martin, N Ramirez, Ysla, (Owens/Johnson)

Some very good K/BB rate pitchers with lower than 7.00 K/9 rates still fair badly with FIP related metrics. To me, low WHIP and low OPS against is way more important than K% or K/9.

What we lose on defense at 1B vs any gain we may get by using Bautista at DH instead of basically a Moreland-Young platoon is not worth the money we pay and the loss of flexibility. What if we sign JB and then we have a major injury at a position of limited depth? With no flex money, we'll end up going over the limit just to get a possible slight gain. I want $6.5M Young playing vs LHPs at DH not riding the bench. I want HanRam at DH vs RHPs not at 1B full time. I don't want a $5.5M Moreland on the bench almost everyday. While I do t2hink JB is better than Moreland vs RHPs, I like Young better vs LHPs, and the loss at 1B defense makes it pretty close to even value to me. I don't see JB as a big plus.

There are too many exceptions to "the rule" that are too obvious to ignore. While a strong correlation does exist between high Ks and good pitchers or as a predictor to future good pitching, it continually misses out on accurately valuing great low K pitchers season after season after season. It's not like a fluke outlier season here and there that don't fit into the formula; it is a large group of pitchers that are continuously short-changed, because they force batters to hit the ball weakly for outs instead of striking out more batters. Look, I get that a guy who Ks a lot of batters compared to the norm is usually a better pitcher than one that Ks less than the norm, but some high K pitchers also let up a lot of hits, HRs or BBs, but their high K totals make them better pitchers using this formula than someone who lets up less hits and BBs per nine, but Ks way less. Just because the fielding portion of a pitcher's performance is removed from the equation, doesn't automatically make it a better formula than one that accounts for pitchers getting more outs and allowing less extra base hits than high K pitchers. Since 1986, these are the top starters by ERA- (2500+ IP) with K/9 rates: 67 Pedro 10.0 69 Clemens 8.6 75 Randy J 10.6 76 Maddux 6.1 76 Halladay 7.0 78 Brown 6.6 78 Oswalt 7.4 79 Felix 8.4 79 Schilling 8.6 80 Saberhagen 6.2 xFIP (K/9) 3.17 Schilling 8.6 3.23 Randy J 10.6 3.23 Halladay 7.0 3.24 Felix H 8.4 3.41 C Carpernter 6.9 3.42 Hammels 8.6 3.44 Pedro 10.0 3.45 J Smoltz 7.9 3.50 Greinke 8.1 3.50 Clemens 8.6 Greg Maddux is 23rd at 3.73 all because of his relatively low K rate. His 1.14 WHIP is third best, but according to this formula, that's secondary to getting Ks instead of ground outs or pop outs. Sorry, I can't drink that kool aid. The worst pitchers in this era according to FIP are: 4.88 Suppan 4.9 K/9 (104 ERA-) 4.82 Trachel 5.74 (101) 4.77 Wakefield 5.9 (97) 4.70 Garland 4.8 (98) 4.65 Williams 5.9 (98) 4.52 B Arroyo 5.8 (98) 4.48 J Moyer 5.4 (97) 4.44 K Rogers 5.1 (94) Really? Wake is the 76th best pitcher out of 79 in this era? All these guys have low K rates, yet they are ranked behind of these guys in FIP: 4.43 B Witt 7.2 (112) 112 for God's sake!!!! 4.38 Astacio 6.8 (105) 4.38 Dempster 7.8 (104) 4.27 Harang 7.1 (104) 4.24 R Wolf 7.0 (101) 4.11 E Santana 7.2 (99) These guys are really better than Wake and Rogers? Bobby Witt had a freakin 1.57 WHIP career! These sample sizes are too large to be outliers or some freaky fluke. They are examples of serious flaws in any system that places too much value on K rates and not on getting hitters to make outs.

Don't underestimate Moreland's defensive value at 1B vs JB's negative defensive value, and the fact that DH'ing HanRam vs RHPs might prevent injury or just keep his bat fresh for the long season. If we hadn't signed Moreland, I could see this debate (or the one for EE previously) as having merit, but I'm with notin. I just don't think DD has any great interest in paying large (even if an underpay) for a DH type. We already have 2 guys who look more like DHs than positional players: HanRam and Pablo. One could argue Young and Swihart fit the mold as well.

You also have to factor in call-ups during the year, of which there are often over 15. Some of these guys go from minor league salaries to minimum major league salaries and those raises count on the budget. Alex Speier projected $4M for this, and he seems to know about these things. If the number is $4M and the arbs come out as projected, then we may have $10-11M in wiggle room remaining.

We all thought he'd return to his norm, which is pretty much what he did or was in the process of doing when the season ended. A hot player (or lucky player) is likely to return to his norm just like a guy in a slump. That doesn't mean a player's "norm" can't change, trend or evolve... most do over their careers, and that's where most predictions rely on age or experience curves to adjust a player's projection based on what has happened to similar players at certain points of their career. It's not an exact science as Papi showed, and sometimes outside influences occur as Barry Bonds showed us. Sandy Leon's career minor league OPS is .654. Within his farm context one notices some horrendous bad stretches and seasons and a couple very high spikes (.856 in 2012 and .938 in winter ball of 2013). His OPS in limited scattered PAs with WSH was .533. It was .439 last year in 128 PAs for the Sox. So, after going through an incredible hot streak (not unprecedented in his professional career), what happened? A .539 OPS in September. That's pretty close to what his career ML OPS was before the hot/lucky streak- whichever you want to call it. Now, the issue is this, because of that hot streak, his career ML OPS has risen from the low .500's to .681. His sample size is still very small at 518 PAs scattered over 5 ML seasons with countless demotions and promotions to and from the farm. What can we expect going forward? For those of you who feel what you were doing most recently counts more, what were you saying last July or August? How far back do you go? Some arbitrary number? If you count just Sandy's last month, you might project a Leon that falls in line with his previous ML OPS and his minor league OPS -between .530 and .630 maybe-- so .580ish? If you like to use a players last 3 months (223 PAs)- you'd think .755 is what we can expect. If you like to use his last season, then .845 might be your number, unless you factor in his .655 at AAA last year in 130 PAs (maybe .775?). To me, I'll stick with projecting somewhere between his career MLB OPS of .681 and his minor league OPS of .654 knowing full well he could have a season much nearer .535 than .845, because that is where most of his career has been. There really was an exact date Leon fell off a cliff-- August 26th. His OPS dropped pretty continuously from the start, but he still was "hitting over his head" for several months: June 1.243 July .956 Aug. .893 Sept .539 He was still over a 1.000 OPS on Aug 26th. Who here could have predicted the correct date for his return to norm? He went 5 for 28 to end August (just 1 double and 1 BB). His September was not below his norm to "compensate" for his hot or lucky streak. It was almost exactly at his norm. Perhaps Leon is a poor example, since his "norm" is not really established, but I used him as an example in response to this post.

They are more likely to return to their norm- not do worse than norm. There are certain hitters who rarely hit for their norm for extended periods. They are always up and down, but predicting the moment they turn their trend is near impossible.

True enough, but I'm hoping we won't be needing top talent... maybe just a de Aza, Ziegler or Leon type that cost us very little. I mention the money, because picking up a so-so player with $5M remaining on his last year of his contract may not cost us top prospects, and we can get two guys like that and still stay under the limit.

Yes, and that's why a .333 hitter has about a .333 chance of getting a hit after going 0 for 20 and a .250 hitter has about a .250 chance of getting a hit after going 10 for 20.

I can flip heads or tails 10 times in a row, but I meant a legitimate coin toss that lands on the ground by a non-skilled magician.

Same as flipping a coin and getting heads 5 times in a row. What's the chance of a heads the next flip? 50%, of course.

It's not .000 either, and it's probably closer to .333 than .000 (higher than .167). That's why I went to a better question. A player's last 20, 50, 100 or 200 PAs is not the best predictor of what is to come in the near or far future.

In the simplest sense, high strikeout rate is a good indicator of a good pitcher. This is the false assumption that irks me. There are countless examples of low K pitchers that had long careers of consistent low ERA- and WHIP. Blaming that on the fielders behind these types of pitchers is missing a whole group of top quality pitchers. Any system that ranks Wright;s 2015 season as below average is seriously flawed. BTW, I saw zero evidence that "proves" this system is a better system than others. Wright's score proves it can't be.