Friday, 23 January 2015

Let's not make a deal.

Always change your guess.

Obviously, sending last year’s 1st-round draft pick, Jeff Hoffman, to the Orioles is a terrible idea.  Terrible, and terribly scary in that Rogers may do it just to make the problem go away, and to hell with the on-field consequences.

Here’s a list of things that might be more appropriate compensation to the Orioles for Dan Duquette.  Or if not appropriate, at least not infuriating:

Any one of Ryan Goins, Maicer Izturis, Danny Valencia or Steve Tolleson (i.e. a Mike Aviles type)

Chase DeJong or Matt Boyd (i.e. a second-tier pitching prospect)

Chad Jenkins (i.e. a Chris Carpenter type)

AJ Jimenez

Kevin Pillar

Coupons for a pregame Steam Whistle tour AND a drunken ride on the adjacent miniature train

The Skywalk

Gregg Zaun, Bob McCown, and Steve Simmons

Last season’s Astroturf.  Baled.

A solemn promise not to type “OrioLOLes” ever, ever, ever again.

Monday, 19 January 2015

The shadow of the Duke

Or... A subreference too far?

It’s tempting to blame the Dan Duquette situation for all the things that aren’t happening with the Blue Jays.

No moves to fix the bullpen or find a second baseman?  It’s because the club doesn’t want to make any transactions under the direction of the lame-duck Beeston.

Arbitration for Donaldson and Valencia?  Nobody wants to make a make a decision and risk tying the hands of the next guy in the President’s seat.

All of that is probably nonsense.   Josh Donaldson may be getting an extension, as we saw happen with Jose Bautista a few years ago when it seemed certain that J-Bau and the Jays were heading to arbitration.  I’d be ok with that, at the right price.  But as a “Super 2”, a player with 4 years of arbitration eligibility, it’s in Donaldson’s interest to get as much money as possible in his first year of eligibility – even more so than would be the case for any run-of-the-mill arb-eligible player.   Arbitration awards tend to snowball from year to year, and the $1.45MM difference between what Donaldson is asking and the Jays are offering for 2015 could amount to as much as – I don’t know, a $15-20MM difference? – over the next 4 years.  So I can see why neither side wants to just split the difference.  Donaldson has been an elite player for the last 2 seasons, and if he puts up another year of 5+ WAR (which would be just fine, thanks) he has the potential to get a $10-15MM arbitration award for 2016.  Lots of money is at stake.

Valencia?  *Shrug* I don’t know why they couldn’t agree.  It looks like the Jays tried to lowball him, and even Valencia’s own proposal was less than what MLBTR had projected for him.  I’d be fine if he wins his case, and if he loses, I don’t care that much if Danny Valencia is pissed off.  He doesn’t figure to be a major piece for the 2016-17-18 Jays like Donaldson should be.

Nonetheless, the ongoing, unresolved Beeston-Duquette situation has fans on edge.  We as fans know that executives don’t draw a lot of compensation when they move from team to team – John Farrell got the Jays a bench piece (Mike Aviles) and Theo Epstein brought Chris Carpenter (this one, not that one) back to the Red Sux.  We know this, but presumably Buster Olney does too, and when Olney says things like one or more of Sanchez, Stroman, Pentecost, Norris or Hoffman would be fair compensation, it makes people nervous.  Surely the Jays wouldn’t be stupid enough to give away anything more than a spare part for Duquette… would they?

You’d hope not.  But then you go out and read other things that make you wonder how rational the people running the show at 1 Blue Jays Way are.  For instance, the Jays might still be interested in James Shields, who is now looking at 4-year deals as opposed to 5-year deals.   And you read that they might be spending $70-80MM to get Shields.  And then you try to reconcile the idea of spending $20MM/yr on a pitcher when we’ve been told that there’s no more than $12 – no, $7MM left to spend.  How does that work?  Deferred money, you say?

Here’s the thing about deferred money… it’s still money.  If the Jays were to sign Shields and pay him $7MM in 2015, they’d have to pay him about $25MM for each of the other three years.   Yes, there is such a thing as time value of money, but it makes very little sense for the club to be okay with a salary of $7MM (but not $20MM) in 2015, when the obligation to the same player would be much, much higher in 2016.   Yes, Mark Buehrle and his $20MM salary are off the books in 2016, but Russell Martin and Josh Donaldson will be getting raises in 2016.  As will Brett Cecil.  Loup and Hutchison will be arb eligible for the first time in 2016.  So it’s not as if the Jays will find 20MM for Shields vastly more affordable in 2016 than they do in 2015.  What the club fetish for salary deferral does is rob Peter to pay Paul, simple as that.  There are no magical savings to be found in 2016 unless you decline options on key players like Bautista, Dickey and Encarnacion.  There are cost savings in 2017, but the team will have huge holes to fill if JB, EE and Dickey aren’t re-signed or extended before then.  Money will have to be found for those spots, too.

I would love to see James Shields in a Jays uniform in 2015 – even if he’s a couple notches below Scherzer and Lester, he’d still be a massive improvement over the guy he’d be replacing, whether you consider that guy to be last year’s 5th starter (Happ) or 2015’s possible 5th starters (Estrada, Norris, or Sanchez).  And all Shields costs is money and a draft pick.  But what I can’t understand is why there might be $80MM available over the next 4 years for Shields, but there doesn’t seem to have been any money available to solve bigger problems (bullpen and 2B) which could have been fixed for half of the $80MM it looks like Shields will cost.

In other words, it would be nice if whoever winds up in the President's chair could insist on a more stable allotment of dollars made available to the Jays.  I'm tired of seeing the financial rug pulled out from under my team, thanks.

Tuesday, 13 January 2015

What to do with Dioner Navarro?

Can't you just picture him fixing your computer?

Immediately after the Russell Martin signing was completed, it looked like Dioner (“Dinner”, “Reboot”) Navarro was a surplus asset.   Martin, as the clear #1 catcher, was going to catch 100+ games, Josh (“T-Hole”) Thole was going to catch Rah Dickey’s 32 starts + the odd day game after a night game, and Navarro would be left with 20-odd games to catch and/or a role as a part time DH.  Protip:  Navarro’s not worth having as a part time DH.    He’s the owner of a .304 career wOBA (.315 last year, projected to .317 this year) and the Jays aren’t going to bench Encarnacion (the regular DH) or any of getting-a-fielding-day-off Bautista/Reyes/Martin to DH Navarro ahead of them.  Even the much-maligned Justin Smoak has a better career (.305) and projected (.327) wOBA than Navarro… and nobody is contemplating DHing Smoak.  Smoak’s value as a player is dependent on a bounceback season and playing a serviceable 1B, not DHing.

And then Martin said he’d like to try catching Dickey’s knuckleball, and the situation kind of changed.  If Martin can catch the knuckler, then T-Hole becomes expendable, and Navarro has a role – his bat doesn’t play at DH, but it plays quite decently at catcher, and it’s a lot more feasible to give Martin days off when his backup isn’t horrible offensively (like Thole is).  On top of that, Navarro is a much more credible option as an everyday catcher if Martin should get hurt for any length of time.  So, 100 games for Martin, 60 for Navarro.  Right?

Well, maybe.  There’s a couple of problems with that idea.  The first one is a practical one – we won’t know how good Martin is at catching knuckleballs until spring training.  If he can’t do it, and the Jays have hung onto Navarro on the expectation that Martin can, the market for Navarro’s services may be a lot smaller than it is now.  Moreover, if you cut Thole loose before spring training, you have no safety net if Martin can’t catch Dickey.

The other problem is opportunity cost.  Dioner Navarro is due to be paid $5MM in 2015.   That’s a lot of money for a backup catcher, money that could be spent elsewhere.  But it's not a lot of money for a starting catcher, and that same $5MM is a good price for a catcher who projects to close to 2 fWAR over a full season as a starter.  In other words, he might be more valuable to someone else who would use him as a starter.   Without the need for a caddy for Dickey, the club could pencil Martin in for 100+ games and give the others to one of A.J. Jimenez or a free agent (personally, I like the idea of taking a shot at Ryan Doumit on a 1-year bounce back deal.  Or Geovany Soto.  Or Laird or Nieves or… you get the idea).  Or hell, even Thole, if $1.75MM isn’t a bit rich for a no-hit backup catcher with the superfluous ability to catch knuckleballs.

What could you get for Navarro?  Beats me.  The Diamondbacks need a catcher and have lots of relief pitchers, so on paper, a trade with Arizona makes sense.  Evan Marshall would be a nice addition to the bullpen, but I’m not sure Navarro gets him, straight up.  Addison Reed?  Maybe, but it doesn’t look like Arizona is willing to give up much for catching help.   Outside of the D-backs, who needs a catcher?  The Braves do, but they just picked up AJ Pierzynski.  Perhaps the White Sox, but they don’t have relievers to deal.   So at the end of the day, even though the Jays need relief help more than they need a $5MM backup catcher, if there are no relievers to be had, the catching flexibility that comes from keeping Navarro may be the outcome the club is stuck with.

Monday, 5 January 2015

Advanced Statistics! What are they good for?

A few weeks ago, I read an interesting article on Blue Jays Plus that covered the philosophical difference between how Fangraphs and Baseball-Reference calculate pitcher WAR (without giving too much away: fWAR is predictive, and rWAR is retrospective). 

Frankly, I hadn't thought too much about this (I know, I know... most people don't think about this stuff at all).  I tend to spend more time at Fangraphs than at BBR lately and I consequently have fallen into the habit of using Fangraphs WAR without considering how it was calculated.  I don't think that's too big of a deal - when I write about pitchers here, it's usually with the intention of postulating how good a player will be next year and making a case for the Jays to acquire/deal a particular pitcher.  fWAR is fine for that.  But it did get me wondering about how the different statistical measures compare, what they do or don't take into account, and what each does best.  It would be really helpful if there was some kind of chart out there that compared the various advanced statistics, as a way of determining which statistics are best used in which situations.

So I looked for that chart.  There are lots of great resources on sabermetric stats online - at Fangraphs, Baseball Prospectus, and other pages - but I didn't find anything like a table comparing each statistic to other stats that measure the same thing.  So, consider this my attempt to compare some of the better-known advanced baseball statistics - what they measure, how they measure it, use of league and park benchmarks, strengths and weaknesses.  I'm sure that I'm missing some stats that could be relevant, but I do hope that I haven't mischaracterized any of the stats I cover below.  If I have, please send me an email and I'll do my best to correct what I've done wrong.  So, here goes: 

On-base plus slugging (OPS)
What is it?  A sum of a player's on-base percentage (OBP) plus his slugging percentage (SLG)
Positionally adjusted? No
Park adjusted? No
Seasonally adjusted?  No.

OPS rates hitters based on a combination of on-base skills and power skills.  On the positive side of the ledger, it's easy to calculate.  On the negative side, it isn't park adjusted or seasonally adjusted, making it hard to compare players who play in different parks or eras.  OPS also gives OBP less weight than SLG, which is inappropriate given that most statisticians feel that OBP is more important than slugging percentage.

What is it? OPS+ is OPS normalized to park and season and set to a base of 100 (league average).  An OPS+ of 110 is 10% better than league average for that season, for example, while an OPS+ of 90 is 10% worse than league average.  Higher OPS+ scores are good for hitters and bad for pitchers (in the form of OPS+ allowed).
Positionally adjusted? No
Park adjusted?  Yes
Seasonally adjusted?  Yes

Better than OPS because it sets everyone up against the same baseline, making it easy to compare a hitter who plays at Safeco with one who plays at Yankee Stadium, for example.  Still doesn't account for OBP being more important than SLG, though.

Runs allowed/9 innings (RA/9)
What is it?  Runs allowed (earned and unearned) per 9 innings pitched
Positionally adjusted? N/A
Park adjusted? No
Seasonally adjusted? No

RA/9 is similar to ERA, but doesn't exclude unearned runs (on the basis that they still count, and a pitcher often has a lot to do with an unearned run scoring, even if the runner reached due to error).  It measures what a pitcher actually did for a team (distinct from FIP and xFIP, below).  RA/9 does not account for park and seasonal effects, but RA+ does (basically in the same way that OPS+ does).  RA/9 is less useful than FIP and xFIP as a predictive tool, because it doesn't account for things like fielding and BABIP.   A few bad bounces, and a good pitcher might have an ugly RA/9 or RA+.

Fielding-independent pitching (FIP)
What is it?  A measure of a pitcher's ability that excludes things he can't control (fielding) and focuses on the things he can:  walks, strikeouts, and home runs.
Positionally adjusted? N/A
Park adjusted? No
Seasonally adjusted? No

The idea behind FIP, as far as I can tell, is to determine how well a pitcher played with respect to things he can control.  A pitcher controls how many hitters he strikes out and walks, and if you think home run per fly ball rate is random, xFIP is FIP that assumes a league-average home run rate.  In theory, FIP is a better predictor of pitcher performance than measures like RA/9 and ERA, because it excludes things that the pitcher can't control.  The problem, though, is that we're not quite sure if things like BABIP and strand rate are outside a pitcher's control, and pitch framing suggests that strikeouts and walks may not be entirely within a pitcher's control, either.  FIP therefore tends to give short shrift to players like RA Dickey, who fields his position well and tends to have a low BABIP due to players getting a lot of weak contact off the knuckleball.

Defense Independent ERA (dERA) is similar to FIP in concept but uses formulae to normalize things like hits allowed, rather than discounting them altogether.

Weighted on-base percentage (wOBA)
What is it?  A substitute for OPS, wOBA weights the different events that get a batter on base (walk, single, double, HR, etc) by the run probability associated with the event.
Positionally adjusted? No
Park adjusted? No
Seasonally adjusted? No

wOBA is very similar to OPS, but adjusted to reflect the consensus that on-base percentage is more important than slugging, and that extra base hits have values that aren't simply multiples of 1 (a double isn't as valuable as two singles, and a HR won't, on average, lead to as many runs as two doubles would, to give a couple of examples).   I am a wOBA adherent; I think it's a very useful tool to help evaluate a batter's ability.  That said, it has its drawbacks:  It's not easy to calculate, the weightings for each on-base event change a bit each year, and it's not adjusted for position, park or season effects.

Weighted Runs above Average (wRAA)
What is it?  A measure of the (notional) offensive runs a player contributed, above the league average.  WRC with league average scaled to 0.
Positionally adjusted? No
Park adjusted? No
Seasonally adjusted? Yes

wRAA, in a nutshell, compares a player's wOBA with the league wOBA, and seasonally adjusts it.  Average is 0, and higher numbers are better (while negative numbers are worse).  A player with more than 20 wRAA is having a good season.  Advantages:  It's seasonally weighted, so you can compare players from different seasons and years.  However, it is not positionally or park weighted, and it doesn't take stolen bases and other non-batting offensive skills (i.e. baserunning) into account.   It also is based on plate appearances, so it's a counting stat.   See also Weighted Runs Created (WRC) and WRC+; WRC also tracks runs created, while WRC+ tracks runs created and accounts for park as well as seasonal effects.

Win Probability Added (WPA)
What is it?  A measure of the change in win expectancy in a game that can be credited to a play made (or not made) by a player.
Positionally adjusted? N/A
Park adjusted? N/A
Seasonally adjusted? N/A

Baseball schedules are very long.  As a result of these long schedules, there have been so many games played that statisticians can determine how probable it is a team will win a game, based on the game situation and how matching game situations in previous games turned out.  Once we know how probable it is for a team to win a game based on a given score/inning/outs/baserunners situation, we can know the value of a subsequent play (the Win Probability Added, or WPA) by comparing the win expectancy of the score/inning/outs/baserunners situation before the play to the score/inning/outs/baserunners situation after the play.  For example, let's assume a team has a 50% chance to win a game, in the 9th inning.  The next hitter hits a home run.  Now the team has a 90% chance to win the game.  The batter who hit the home run has 0.4 WPA from that play and the pitcher who gave up the home run has -0.4 WPA from the same play.  If the batter had instead struck out, his team might then have a 45% chance to win the game.  The batter would receive -0.05 WPA from the strikeout and the pitcher would receive 0.05 WPA.  Over a season, an average player would have 0 cumulative WPA.  Good players would be in the 3+ range and terrible players would be below -1.  In this way, WPA looks superficially similar to WAR.

WPA measures what a player has done, so it can't be used as a predictive stat.  It also doesn't take defense into account.  And it also makes assumptions that may not make sense - for example, it assumes all teams start with a 50:50 chance of winning, even when last-place teams play first-place teams.  What it does do is reflect situational performance - i.e. "clutch" play.  A 2-run HR in the first inning of a scoreless game might add 0.2 or 0.3 WPA, but a 2-run HR in the bottom of the 9th inning of a 5-4 game would likely add 0.9 WPA or more (turning an almost-certain loss into a win)

What is it?  A meaure of defensive ability based on play-by-play data and expressed as "Runs Saved".  League average is 0, higher scores are better, lower are worse.  Anyone above 15 (good) and below -15 (bad) is exceptional.
Positionally adjusted? No
Park adjusted? No
Seasonally adjusted? No - but unlike UZR (below), TZ has data going back past 2002, and thus can be used to compare current fielders with historical players.

I wish I was capable of quickly explaining how TZ is calculated.  The method of calculation varies with the data available (there's more data in more recent seasons).  A fielder is charged with hits based on the location of the hit.  Rate of hits is compared to league averages, converted to runs, and then converted to runs saved above (or below) average.  TZ is considered less accurate than UZR but as noted, it has more utility when making historical comparisons.  Note that the general consensus is that you really need 3 years of fielding data to draw a conclusion about a player's defensive ability

What is it?  A measure of defensive ability based on range, errors, double plays, throwing arm, and other fielding elements.  League average is 0, higher is better, lower is worse.
Positionally adjusted? No
Park adjusted? Yes
Seasonally adjusted? No

UZR is considered more accurate than TZ because more factors go into the calculation of a player's rating.  That said, a lot of these factors are subjective.  Data used in UZR is compiled by human observers, who have to determine whether a ball is hit slow, medium or fast, whether it is a fly ball or a line drive, whether it was an easy play or a hard play, whether a shift was in effect, etc.  One point in UZR's favour is that it is park adjusted, so good defenders who play in spacious, hard-to-defend-in parks get credit for doing so.

Defensive Runs Saved (DRS) (See
What is it?  Another defensive metric which calculates runs saved based on whether a player should or should not have made a play.  For example, if a player makes a play that is only successful 8% of the time, he gets .92 points.  If he fails to make a play that is made 80% of the time, he loses .8 points.  League average is 0, higher is better, lower is worse.
Positionally adjusted? No
Park adjusted? No
Seasonally adjusted? No

Like UZR and TZ, you should look at 3 years of data to have a clear idea of a player's defensive skill.  Also like UZR, DRS relies on human scorers to compile data, so the results may be a bit subjective, especially in small samples.  UZR and DRS tend to produce similar scores.

Put it all together, and what do you get?


WAR (Wins Above Replacement) is calculated by taking the estimated number of runs a player produces (or saves) through batting, defense, and baserunning (and for pitchers, pitching), and dividing by 10 (10 runs are considered to equal about one win).  Note that a replacement level player (a AAA call-up, for example) is not the same as a league-average player; a league average player is worth between 1 and 2 WAR.  A player who is worth 3 WAR is about 30 runs better than a replacement level player, and 12-18 runs better than an average player. and each have their own versions of WAR, with slight differences in how they calculate position player WAR and significant differences in how they calculate pitcher WAR.  Baseball Prospectus has another version, called WARP, which I will ignore for the purposes of this post.

rWAR (position players/DH's)
What is it?  Baseball-reference's version of WAR (Wins Above Replacement), also called bWAR.  A measure of a player's offensive and defensive ability, using wRAA for batting, Total Zone rating for defense, and a separate baserunning measure.
Positionally adjusted? Yes (+10 runs for a catcher, −10 for a first baseman, +3 for a second baseman, +2 for a third baseman, +7.5 for a shortstop, −7.5 for a left fielder, +2.5 for a center fielder, −7.5 for a right fielder, and −15 for a DH)
Park adjusted? Yes
Seasonally adjusted? No

fWAR (position players/DH's)
What is it?  Fangraphs' version of WAR.  A measure of a player's offensive and defensive ability, using wRAA for batting, UZR for defense, and something called Ultimate Base Running (UBR) for baserunning.
Positionally adjusted? Yes (+12.5 runs for a catcher, −12.5 for a first baseman, +2.5 for a second or third baseman, +7.5 for a shortstop, −7.5 for a left fielder, +2.5 for a center fielder, −7.5 for a right fielder, and −17.5 for a designated hitter)
Park adjusted? Yes
Seasonally adjusted? No

The two versions of batting WAR take essentially similar approaches, and in fact, both use WRAA to measure batting skills.  However, they each use different baserunning and defensive measures, and apply different positional adjustments to runs created/saved.

rWAR (pitchers)
What is it?  Baseball-reference's version of WAR for pitchers, calculated using Runs Allowed (RA) and Innings Pitched.
Positionally adjusted? N/A
Park adjusted? Yes
Seasonally adjusted? No

fWAR (pitchers)
What is it?  Fangraphs' version of pitcher WAR, calculated using FIP and innings pitched.
Positionally adjusted? N/A
Park adjusted? Yes
Seasonally adjusted? No

The two versions of pitcher WAR take radically different approaches to determining pitcher value:  rWAR looks at what a pitcher did in a given year, while fWAR looks at what a pitcher should have done.  Taken another way, you should look at rWAR when picking a Cy Young winner, and fWAR to determine who is more likely to pitch well next season.  In both cases, more innings equals more value.

And that's all I've got.  Sorry about the wall of words, but even if you knew all this stuff already, I hope it helps to have all this information down in one place, anyway.

Thursday, 25 December 2014

A short rant for the season

At times, the way the current Blue Jays management does things drives me crazy.

Case in point:  Dan Duquette.

Dan Duquette may well be a very good baseball GM - he deserves a lot of credit for the players the Expos developed in the early 90s, some credit for the Red Sox resurgence in the mid-2000s, and to be fair, not a whole lot of credit for the Orioles recent successes.  Despite this, it would be idiotic to surrender any kind of cost-controlled talent in order to retain Duquette's services.  I get that the Jays need - or want - to replace Paul Beeston.  I also get that they like Duquette.  And Kenny Williams.  And who knows who else.  But no one front-office type has ever been worth a Stroman, Norris, Hoffman, or any other top talent.  Publicly fixating on Duquette the way Rogers has done will only drive his price higher, and besides, the Toronto sports landscape is littered with the bodies of saviour front-office suits (Burke, Leiweke, Colangelo) who either failed or skipped town.  The Jays would be wise not to bet the farm on Duquette... but I worry that Rogers may adopt a Duquette-at-all-costs approach.

Beyond the search for Beeston's replacement that is rapidly becoming embarrassing, there's the issue of (re)stocking the bullpen.  The Blue Jays, and AA, have given every indication that they prefer to do this via the trade market, as opposed to signing free agents.  Why?  I have no idea.  If the Jays want to get a Greg Holland type, theyll need to give up something of value for him - something cost-controlled for a few more years.  In return, they'll get a player who is already making almost $5MM per year, is due a raise for 2015, and who will be a free agent in 2 years.  OR - they could just pay for a reliever, get him for 3-4 years, and not have to give up anything.

Trading for relievers - who are volatile and whose value is effectively capped by the number of innings they throw - costs players, money, and seasons of control.  Free agents just cost money.   Trading for Josh Donaldson made sense because a player of Josh Donaldson's calibre wasn't available through free agency.  Trading for a Papelbon or Holland makes no sense, because you could get a Robertson or Miller for more years (and roughly the same $/yr) without giving up any prospects.

Why is this important?  Well, for the last several years, the Jays have been getting most of their on-field performance from high-priced players that were brought in from elsewhere - Bautista, Reyes, Buehrle, Dickey, Encarnacion, Rasmus, etc.  The team has done very well to have players like Bautista and Encarnacion fall into its lap, but at some point, the club needs to develop some elite players internally.  Stroman is a good start, and guys like Sanchez, Hutchison, and Pompey may get there too.  Trading away top prospects for short-term relief help, or fungible front office types, is not the way to keep the talent pipeline flowing.

Happy Festivus.

Monday, 15 December 2014

The Melkman goeth.

The writing had been on the wall ever since the Jays traded for Michael Saunders, but now Melky Cabrera is officially gone, having signed with the White Sox for 3 years, $42MM.

You can go back and forth over whether Saunders is a downgrade from Cabrera.  It comes down to this:  Saunders is a better defender and baserunner, and Melky is a better hitter.  In the end, I'd say Cabrera is more valuable - even if they have produced (and project to) comparable WAR totals, Saunders gets a lot of his value from defense, and defense seems (to me, anyway) to be a bit over-valued by WAR.  It's not a landslide for Melky - but it's a tangible downgrade. 

On top of the head-to-head comparison between the two left fielders, you have to consider Saunders' health, his cheaper salary, the fact that Melky is on the wrong side of 30, having to give up the not-totally-useless Jah Happ to get Saunders, and whether the Jays will spend the $10MM difference between the two salaries on much-needed pitching help.   What Melky's signing does do is provide an excuse to go over how compensatory draft picks work.  So, in case you're interested, here goes:

1.  A prospective free agent player who is given a qualifying offer (1 year, at the average dollar value of the 125 richest contracts, about 1 yr/$15MM) by his former team has the option to take the qualifying offer (QO) or decline it.  If he takes it, he returns to his former team at that salary.  If he declines, he becomes a free agent, and his former team gets a compensatory draft pick if the player signs with a new team before the next season's draft day (the following June).

2.  When the free agent that turned down the QO signs with another team, that team forfeits its highest unprotected draft pick in the following year's draft.  The top 10 picks in the draft are "protected" and can't be lost in this way.  Compensatory picks from free agents who signed elsewhere are not protected.

3.  The compensatory pick that a team that loses a free agent receives falls in a round between the first and second round of the draft.  The pick that was forfeited by the team that signed the free agent has no bearing on the positioning of the compensatory pick.  If multiple teams hold compensatory picks, they pick in the same order they would pick in the draft (i.e. in reverse order of finish).  Compensatory picks cannot be traded.

4.  The order in which free agent signings happen is, as far as I can tell, irrelevant.  If a team forfeits a first-round pick and a second-round pick due to signing QO'd free agents and then one of its own QO'd players signs elsewhere, the team gets its second round pick back, but forfeits the compensatory pick, instead.

So:  The Jays lost the 18th pick in the draft by signing Russell Martin.  This pick doesn't go to the Pirates, or anywhere else - it 'disappears', and everyone below the Jays in the draft order moves up 1 spot.  The Pirates get a pick in the compensatory round.  The White Sox then signed Melky Cabrera.   The White Sox hold the 8th pick in the first round, which is protected, so instead, Chicago loses its highest un-protected pick (their second-round pick).  As when the Jays lost their #18 pick, this pick vanishes.  The Jays get a pick in the compensatory round for 'losing' Cabrera.  As it turns out, the Jays compensatory pick is currently the #31 pick in the draft, while the Pirates compensatory pick is #33.  This difference is due to the Jays having a worse record than the Pirates in 2014 - the worse record picks higher.

If more teams lose picks in the first round due to signing free agents, the Jays' compensatory pick would move up one spot for every pick forfeited ahead of them.

If the Jays sign another free agent that received a QO (say, James Shields), they forfeit the compensatory pick they received for losing Cabrera.

Now that I have all that figured out, I hope MLB doesn't go changing the system again for a few years...

Friday, 12 December 2014

Post-winter meetings thoughts

The Jays didn't do anything at the winter meetings.  This is a cause of garment-rending in some parts... but before we start jumping off buildings in despair, let's remember that it's still just December, the Jays have made lots of moves already (upgrading at C and 3B and finding a LF who may be as good as Melky if you consider both offense and defense), and, most importantly, there are still decent players left to be had.  Such as (in descending order of how interesting they are)...

Takashi Toritani, 2nd base:  The hot rumour of late is that the Blue Jays are very interested in Toritani.  He's a 33 year old shortstop who has compiled an impressive record of durability, on-base skills (over .400 the last 2 seasons) and defense, in Japan.  He'd be looking at about 3 yrs/10MM to come play here.  I like the idea of picking up a player who gets on base to play in the middle infield, and Toritani has shown doubles power in Japan as well.  He would play 2B in Toronto, but could presumably play some shortstop when Reyes is DHing (and someone like Valencia or Izturis could play 2B).  Signing Toritani would also make the infield deeper (Goins/Travis would be in Buffalo).  Toritani would come cheaper than a Jed Lowrie type, too.  What's not to like?

Burke Badenhop, reliever:   A sinkerballing reliever who generates a ton of ground balls, Badenhop has strung together 3 pretty good seasons.  He doesn't walk many and doesn't strike out many, but he produces.  Could be a fit with a Jays team which has improved its infield defense and still plays in a stadium that punishes flyball pitchers.

Casey Janssen, reliever:  Hey, why not?  The Jays know Janssen, he knows them, and if his second-half issues that may have torpedoed his value were all food-poisoning based, he could be a bargain at, say, $4MM/yr. 

Mike Adams, reliever: Hey, remember how good Mike Adams was a couple of years ago?  Well, he had shoulder surgery in 2013, was pitching well on his return from that last year, and then went on the DL again with shoulder inflammation.  He pitched in September without any issues.  I wouldn't build a bullpen around him, but he could be a decent buy-low candidate.

Jason Motte, reliever:  Another rebound candidate, Motte struggled in 2014 after 2013 surgery, but was very effective from 2010-2012.  And he's just 32.

You could also take flyers on guys like Ronald Belisario and Chris Perez to staff the bullpen, or a "proven closer" like Jason Grilli or Sergio Romo.  Even without those names, nabbing 2 or 3 of Badenhop/Janssen/Adams/Motte would provide a fair bit of depth and upside to the 'pen on their own.

Mike Morse, 1st base:  Hey, he's probably better than Smoak... at least with the bat.  Career .359 wOBA (.355 last year) sounds good. Yes, he's righthanded, but he has virtually no platoon split over his career.  He will cost a bit more, though... and his defense is reminiscent of Edwin Encarnacion's.  That's not good.

Kris Medlen, starting pitcher:  Medlen didn't pitch at all in 2014 due to an elbow injury that required surgery.  If the Jays don't want to spend the money or prospects to get a mid-rotation or better starter, they could do worse than to take a flyer on Medlen, who provided 6.3 fWAR of value from 2012-13.

And that's just the free agents.  For the sake of the future, I'd rather see AA pick up some free agents as opposed to dealing upper-tier prospects.  If it doesn't go that way... well, I won't even try to guess what kind of trades might be out there.

Wednesday, 3 December 2014

Happ, Dirks, Mayberry gone, Michael Saunders here, Smoak gone and here again...

F***!  Is that Astroturf?

Funny thing about writing a sports blog as a hobby - sometimes, events move faster than I can write.

Last night - Tuesday night - the Jays non-tendered Justin Smoak, Andy Dirks, and John Mayberry, and tendered a contract to Marco Estrada.  Let's talk about that last point, first.

I had speculated a while back that Marco Estrada might be non-tendered... and he wasn't.  Estrada's going to wind up costing about $4MM in arbitration, which seemed excessive.  I say "seemed" because the hot rumour now is that top free-agent non-closing relievers like Andrew Miller might cost 4 years and $32+MM.  Right now, the Jays' bullpen is Brett Cecil (who was tendered), Aaron Loup, Todd Redmond, Marco Estrada and maybe Aaron Sanchez.  Subtract Estrada from that mix and you probably can't consider sending Sanchez to Buffalo as a starter - you'd need him for the bullpen.  And in related news, Casey Janssen would be open to talking with the Jays about a return to Toronto.  That'd help, too, at the right (Estrada-sized) price, but even with Janssen back in the fold, I think the Jays would still need another bullpen arm if they plan to send Sanchez to Buffalo.

And then the Jays traded Jah Happ to Seattle for Michael Saunders, which makes Estrada more than a bullpen option, and makes that $4MM arbitration number look better.  If Sanchez doesn't break camp with the Jays in March, Estrada is the 5th starter, as things stand right now.  Hopefully, it doesn't come to that - Estrada's numbers are much better as a reliever than as a starter, and if the Jays are going to have to overspend on pitching, I'd rather they throw money at a starter - if not a Lester, then a Jason Hammel/Justin Masterson type.  Buying 2 relievers with Estrada as a weak 5th starter doesn't sound nearly as good.

As for Happ - the Jays sold high on him (for a change), after Happ cut his walk rate per 9 to a career-low 2.91 and found an extra couple mph on his fastball, en route to what was arguably his best season since 2009.  Happ is probably better than Estrada, and maybe better than we can reasonably hope Sanchez would be as a rookie starter.  But trading Happ saves the Blue Jays some cash - depending on how you look at it, the Jays save $4MM (the difference between Happ's salary and Saunders' salary) or somewhere in the $10MM range (the difference between Saunders' salary and what a free agent LF would cost).

So, Saunders.  In exchange for Happ (under control for 2015 only), the Jays get a 28-year old outfielder with 2 years of control left.  Saunders is lefthanded (filling a need for Toronto), plays good defense in LF, and can hit more than we would expect from Dirks/Mayberry/Pillar - but probably not as well as Melky Cabrera would.  He's a better defender than Melky, a better base-stealer, and obviously, a lot cheaper.  Saunders missed half of 2014 due to injury and consequently fell out of favour with Seattle management, but that's the only major knock against him.

So, what does this mean?  Well, the Jays now have a hole in the back of their rotation instead of in left field.  They have a cost-controlled LF who figures to provide almost as much value as Melky Cabrera did in 2014. They have a few million bucks more to play with.  And they have a draft pick, if and when Melky signs with someone else.  I suppose there is a slight chance they put Saunders in CF (where he isn't very good, fyi) and bring Melky or another LF back, but I doubt it.

Oh, I almost forgot - the Jays re-signed Justin Smoak for $1MM.  As I see it, he's still penciled in for about 300 PA at first base, with the other half going to Edwin Encarnacion.  While Edwin plays 1B, Reyes, Bautista, and others will rotate through DH.

Expect more moves to come.  The Jays still need a second baseman and 2 pitchers (2 relievers or a starter and reliever) and over the past two days, they've freed up about $10MM to go and get them with.

Friday, 28 November 2014


OK, that's some kind of trade.

If you don't follow the Oakland A's closely, you may not know what kind of player Josh Donaldson is.  Be assured that he is a very good player.  According to Fangraphs and Baseball-reference, he's been a better player than Bautista over the past 2 seasons, which are his only seasons as a full-time player.  Donaldson fields his position well, hits for power (20+ HR each of the past 2 seasons) and draws walks (76 in both 2013 and 2014).

Want more?  He's played in all but 8 games over the past 2 seasons.  He's about as injury-free as they come.

More?  He's under control for 4 more seasons.  Donaldson is a so-called "Super 2" player, which makes him eligible for arbitration before this season.  So he'll be getting a raise, but not a backbreaking one, one would think.  My guess is that the Jays try to buy out his arb years and a couple of free agency seasons, but those considerations can wait a bit.

In brief, Donaldson is a middle-of-the-order bat who will give the Jays great defense, and hopefully better durability, at the hot corner.

And the Jays made this trade by giving up just Brett Lawrie - another controllable (if that's ever the right word to describe Lawrie) third baseman, someone who may one day be a great player, but who is clearly not in Donaldson's class, yet - and three prospects.  Franklin Barreto is the Jays' 8th-best prospect according to, Sean Nolin's the #11 prospect, and Kendall Graveman is admittedly a bit hard to rank.  Making a deal like this without giving up the team's best prospects is impressive.

It stings a bit to give up Brett Lawrie, a guy who oozes talent and energy but hasn't been able to put everything together over a full, healthy season.  It may surprise you to learn that Lawrie will actually be eligible for free agency a year before Donaldson will.

My guess is that the A's have made this trade because they don't want to pay Donaldson over the medium to long term, and they figure that they are getting a player in Lawrie who may someday be almost as good as Donaldson, plus 3 not-bad prospects who might help at the plate (Barreto isn't considered a good enough defender to stick at shortstop) and in the rotation in a couple of years.

Meanwhile, the Jays have added a premium player - one who is clearly better than Pablo Sandoval, incidentally - for their incumbent at the same position and three middling-to-good prospects.  And nowhere near Sandoval money.

And I'll say it again:  The Jays pulled off this trade without giving up any of their top young players/prospects (i.e. Stroman, Sanchez, Pompey, Norris)

Exciting stuff.

Thursday, 20 November 2014

Pipe dreams

Rob Ford jokes are already passe.  Sorry.

So... rumours are swirling around our Toronto Blue Jays.  Some exciting, some troubling.

According to Bob McCown via (if you haven't figured out that Drunk Jays Fans is now Stoeten's eponymous site, that's the scoop.  Go check it out), the top brass at Rogers are looking to repair the company's reputation by not being liars   having a consistent approach to the team budget  building a winner out of the Jays.

Meanwhile, there are other rumours (from Jeff Blair via The Blue Jay Hunter) that suggest that Rah Dickey might not be the best teammate out there.  Considering that there were rumours about Rasmus's aloofness, and Adam Lind seemed to complain a lot, and those two are both gone or as good as gone... Dickey might be on the outs as well.  It's a little disturbing to think that the guy the Jays built their 2012 renaissance around (and who they traded two top-50 prospects for) might be persona non grata, only 2 years later.

But nevermind that - it's daydreaming time, and there might be money to spend!  Put those two points together, and let's look at some deals that would have seemed impossible 2 weeks ago.

1.  Jon Lester to Blue Jays. 


No, seriously, it could kind of work.

Two weeks ago, "Bahaha" would have been the end of this post.  But if there is anything to Bob McCown's image-burnishing rumour, maybe the Jays could be in on Jon Lester and other expensive free agents.  Which would be pretty exciting, to say the least.

Jon Lester has a lot of things going for him.  He's pitched in the AL East, obviously.  He's been durable - 30+ starts for 7 seasons.  He's been effective - 33 fWAR in those same 7 seasons.  And while his strikeout rate was on a slow decline prior to a resurgence last season, his walk rate has been dropping as well.  The downside for Lester is his age, and the years he wants.  He'll be 31 next season, and is apparently looking for a 6-year deal, at around $20MM/season (the Red Sox reportedly offered him 5/$100MM to come back, already).

So how could this work for the Jays?  Well, they'd have to count on Lester being effective into his mid-30s, the way they're counting on Russell Martin to do the same, and the way Mark Buehrle has.   Yes, it's a lot of money, but not that much if you can subtract Dickey's salary after signing Lester.

Trade Dickey?  Sure.  Dickey and Lester have carried similar workloads over the past 4 years, so Lester could effectively replace Dickey's innings.  Lester also looks like a fairly significant upgrade to Dickey, by fWAR.  And if you trade Dickey, you free up about $12MM for the next 2 years (assuming Dickey's 2016 option gets picked up) and you also free up a roster spot that won't have to go to Josh Thole.  And finally, Dickey has some trade value attached to him - at $12MM/yr, he's a relative bargain for a pitcher who is capable of throwing 200+ above-average innings, and he could bring something half-decent back in trade.  So, yeah, swapping Dickey out for Lester could be a big upgrade, at a net salary increase of only $8MM or so (until we get to 2017, but let's pretend not to think about that).

2.  Pablo Sandoval to Blue Jays.

No, no, no. 

I think Sandoval is a pretty fun player to watch, but he's another guy looking for 5 years and close to 20MM/yr.  And consider this:

Sandoval, 2011-2014:  13.4 fWAR,  .383/.338/.331/.323 wOBAs
Chase Headley, 2011-2014: 17.9 fWAR, .340/.378/.330/.316 wOBAs... and better defense.

Yes, Headley is 30, and Sandoval is 28.  Headley is an injury risk, but I'd argue that when playing on Astroturf, the 5'11, 250-lb Sandoval is too.  So if Sandoval wants 5 years, offer 3 years and the same AAV to Headley.  I bet he signs. 

And for that matter, if you sign Jed Lowrie or Stephen Drew, both of who are expected to go for 3yrs/30MM or so, you can keep Brett Lawrie at his preferred position (and his best defensive position).  Best case for everyone?  The Panda goes back to San Francisco.

3.  Hanley Ramirez to Blue Jays.

Same objections as Sandoval, really, with the added negative of having to ask him to move off shortstop (or ask Reyes to move, which could be just as big of an issue).  12.3 fWAR and .317/.328/.442 (small sample)/.362 wOBAs over the last 4 years.  Better bat than either Sandoval or Headley, but a suspect glove, and more injury prone than either one.  Seriously, just get one of Lowrie/Drew/Headley.

4.       Andrew Miller to Blue Jays.

Well, the performance of relief pitchers is supposed to be volatile (therefore making them risky investments) and the Jays made a point over the last few years of acquiring cheap bullpen arms (Delabar, Santos)… who didn’t work out.  And now the Jays, having said goodbye to Casey Janssen and Dustin McGowan, may have come full circle and are apparently in the market to overpay for a reliever.

The first rumour I saw was Francisco Rodriguez to the Blue Jays.  Kinda makes sense on the surface - the man once known as K-Rod is a so-called "proven closer", coming off a 44-save season in Milwaukee.  Trouble is, his peripherals are lousy:  declining strikeout rate, tons of home runs (1.85 per 9 innings), an unsustainably low BABIP (.216) and unsustainably high strand rate (93%).  In short, Rodriguez's stats scream "Regression candidate!"  So, overpaying for that closer seems like a bad idea.

Andrew Miller, on the other hand, has been pretty awesome.  Since being made a full-time reliever in 2012, he's posted huge K rates, unpleasantly high walk rates that he seemed to get under control in 2014, and ERAs that aren't luck driven.  He doesn't have significant splits between righties and lefties.  He's the best free agent reliever out there who isn't David Robertson.  MLBTradeRumors guesses that Miller will get a 4-year, 32MM deal.  That's closer money - no, it might actually be starter money.  Miller has been worth 3.5 fWAR over the last 4 year (including the 2.3 fWAR he produced last year).  4yrs/$32MM is an overpay at that performance level.  In fact, Miller's 4 best years for his career only add up to 5.2 fWAR.   Sure, he was great last year, but there's a good chance that he winds up a 1 WAR/yr kind of guy.  If that happens, the only way to justify it is with a lot of saves.

So that's 4 dreams, 3 of them bad.  But at the end of the day... it's not my money.  So go ahead and try to buy our love, Rogers.