Blue Jays Discussion: No longer the off-season. It's time for real baseball

  • Xenforo Cloud has upgraded us to version 2.3.6. Please report any issues you experience.
Status
Not open for further replies.
Roark was pretty good in his first 3 outings. Obviously the stinkers are going to be there and we're going to have to live with it. I'd go with Kay as the 5th guy for now if it were up to me, but Roark is going to have to lose that spot in the regular season first.

I don't mean to toss out his other starts just because of this one but the earlier a start is in the spring the greater the chance that it was against some less than stellar competition.

Honeslty, I'd give Manoah some early run. See if his performance this spring is real. If it is, awesome, he basically dares the team to send him down when Pearson gets healthy. If not, no big loss and you can switch him out for Zeuch, Kay, Thornton, or anyone else who isn't Roark (I'm presuming in this scenario that Stripling is the #4 until Pearson is back)
 
  • Like
Reactions: kb and Puckstuff
I don't mean to toss out his other starts just because of this one but the earlier a start is in the spring the greater the chance that it was against some less than stellar competition.

Honeslty, I'd give Manoah some early run. See if his performance this spring is real. If it is, awesome, he basically dares the team to send him down when Pearson gets healthy. If not, no big loss and you can switch him out for Zeuch, Kay, Thornton, or anyone else who isn't Roark (I'm presuming in this scenario that Stripling is the #4 until Pearson is back)
Only issue with that is that you start the service time clock. But wow, has Manoah been dominant this spring. Too bad he already got sent out.
 
Only issue with that is that you start the service time clock. But wow, has Manoah been dominant this spring. Too bad he already got sent out.

Baseball is coming closer and closer to closing the service time manipulation loophole that at this point I think you might as well just go for it. Also because as good as Manoah has been he's not of the Vlad/Bichette/Pearson tier where the expectation is that they'll break the bank sooner rather than later, so perhaps there's some "value" to be gained in showing you won't play those games so much anymore before it gets to the point that MLB and the PA force teams' hands.
 
Baseball is coming closer and closer to closing the service time manipulation loophole that at this point I think you might as well just go for it. Also because as good as Manoah has been he's not of the Vlad/Bichette/Pearson tier where the expectation is that they'll break the bank sooner rather than later, so perhaps there's some "value" to be gained in showing you won't play those games so much anymore before it gets to the point that MLB and the PA force teams' hands.
In all honesty, we don't know the tier of prospect he is. He's pitched a total of 22 innings since being drafted - including the 5 this spring. His pro stat line is ridiculous.

22 IP, 14H, 5R, 5ER, 5W, 38K

I mean.....that's Pearson+
 
apparently Palacios' elbows are made of wood. How else does an umpire hear him getting hit in the arm and think it's bat?
Maybe you're right, do you mind explaining it to me in simple terms?


If we left the explanation up to Tabby you'd probably be more confused by the end, not less.
 
The Rockies have been crazy for years now... that really doesn't surprise me all that much.
 
Pretty much guarantees Cole will join the team with Liriano who was likely to make the roster;

CL - Romano
SU - Dolis
Phelps
Chatwood
Merryweather
Borucki
Liriano
Cole
Thornto/Kay

Still not a bad assemble of arms. Shane Greene is still a free agent. Wonder if Cashner or Robertson are healthy.

He hasnt pitched in 2 years so i doubt he is healthy but so is Brett Cecil.
 
Pretty much guarantees Cole will join the team with Liriano who was likely to make the roster;

CL - Romano
SU - Dolis
Phelps
Chatwood
Merryweather
Borucki
Liriano
Cole
Thornto/Kay

Still not a bad assemble of arms. Shane Greene is still a free agent. Wonder if Cashner or Robertson are healthy.

He hasnt pitched in 2 years so i doubt he is healthy but so is Brett Cecil.

Merryweather also hasn't pitched in a game yet, so I would be surprised if he's ready to go for opening day. Thornton and Kay both getting a spot would be fine with me.

I'm not big on giving Liriano the second lefty spot... Bergen is already on the 40-man and having a strong spring, and if they're going to open up a spot I would prefer that it go to Mayza. I think Liriano will get it, but it's probably my third favourite choice for that spot.

Cole's kind of the same for me... he's fine, but there are more interesting arms performing well enough that I would prefer to see on the team.

Oh well. Lots of decisions left to be made all over the roster... should be an interesting last couple of weeks.
 
Merryweather also hasn't pitched in a game yet, so I would be surprised if he's ready to go for opening day. Thornton and Kay both getting a spot would be fine with me.

I'm not big on giving Liriano the second lefty spot... Bergen is already on the 40-man and having a strong spring, and if they're going to open up a spot I would prefer that it go to Mayza. I think Liriano will get it, but it's probably my third favourite choice for that spot.

Cole's kind of the same for me... he's fine, but there are more interesting arms performing well enough that I would prefer to see on the team.

Oh well. Lots of decisions left to be made all over the roster... should be an interesting last couple of weeks.

Didnt realize what Bergen and Mayza were doing in comparison to Liriano. Also didnt realize Merryweather isn pitching.

Maybe the Jays add Liriano and a 3rd lefty like a Bergen.
 
  • Like
Reactions: Discoverer
Maybe you're right, do you mind explaining it to me in simple terms?

Sorry, I would've explained it at the time, but I thought we were just making fun of Tabby, not that it was an actual question.

WAR is "Wins Above Replacement"

One important caveat is that while it is a stat when you pull it up from a given website, the overall concept of WAR is just that: a concept. There are multiple stats under the umbrella that measure the same idea in slightly different ways for slightly different reasons. But we'll get there later if you want to read all of that.

The short, short, short version is that it is an expression of the value provided by a player in all facets he can contribute with (with the bat, in the field, and on the basepaths for position players, on the mound and a tiny bit with the glove for pitchers. NL pitchers' hitting is such a small amount that it's pretty negligible) to his team's success, done on a scale that allows you to normalize any variations in how the league performs as a whole from year-to-year or era-to-era.

If you just want to be able to use it to judge player value, Fangraphs suggests the following brakedown as a good rule of thumb (and as a side note, whenever I'm using WAR here I'm referring to the numbers on Fangraphs. This is important, but I'll get to the "why" of that later)

<0 = probably shouldn't be playing in MLB
0-1 = Scrub
1-2 = Role player
2-3 = Solid starter
3-4 = Good player
4-5 = All-Star
5-6 = Superstar
6+ = Potential MVP candidate

A little more depth and some examples using the 2019 season (and I'm going to start using spoiler tags because otherwise this post will be HUUUUUUUUGE):

For comparison's sake, in 2019 (because 2020 wasn't a full season) your position player leaders in WAR according to Fangraphs were Alex Bregman and Mike Trout at 8.5. George Springer was in the top 10 at 6.5

The pitching leader was Gerrit Cole at 7.3 Other guys in that range were Jacob deGrom at 7.0. Max Scherzer at 6.5 or Justin Verlander at 6.3.

The 5-6 range had names like DJ LeMahiu, Ronald Acuna and JT Realmuto. Hyun-Jin Ryu wrapped up his last pre-Blue Jays season at 5.0. Shane Bieber and Stephen Strasburg could also be found here.

4-5 had names like Bryce Harper, Anthony Rizzo, Nelson Cruz, and Michael Brantley. Mike Soroka, Kyle Hendricks, Luis Castillo, and Zach Weeler were in this tier.

3-4 had guys like Justin Turner, Yuli Gurriel, JD Martinez, and Corey Seager. On the mound there was Aaron Nola, Matt Boyd, Jose Quintana, or Clayton Kershaw's down year.

2-3 was where you found Marcell Ozuna, Mike Moustakas, Kyle Schwarber, and Adam Frazier. your 2019 pitchers at this level were Robbie Ray, Anibal Sanchez, Yu Darvish's somewhat disappointing turn with the Cubs, or Tanner Roark's last OK season before he went to the Jays and imploded.

1-2 was guys like Dexter Fowler, Elvis Andrus, Kevin Pillar, and Freddy Galvis. Pitchers taht finished around here included Mike Leake, Mike Fiers, and the disappointing implosions of Jeff Samardzija and Julio Teheran.

0-1 had guys like Rougned Odor, Brandon Belt, Ranal Grichuk's bad year, or Colin Moran. No qualified pitcher finished with a WAR below 1.0 last season. If I set the bar as low as 130 innings for the season then we get Aaron Sanchez's last turn with the Jays and his Astros debut which logged 0.8 WAR

and your negative WAR leader was Khris Davis in Oakland at -0.9 on account of being a pretty terrible bat and very poor defender who still played 130+ games. Using the same 130 IP threshold I did above, the only negative WAR regular was Glenn Sparkman for Kansas City, who finished at -0.5


Of course those examples are just single season snapshots. Pillar has had seasons where he was better than that, as have guys like Turner, Seager, Moustakas, Harper, etc. It's just how they were in 2019



If you want a little bit more depth, you can read on:

What is "Replacement-Level" really?
The replacement-level player is a hypothetical player that is freely available to acquire at no cost other than a league-minimum salary. So it's guys that are MLB free agents kicking around the minors, nobodies you might pluck off the waiver wire, or non-prospect org depth type players you could call up from your farm system. I think sometimes people believe that replacement-level means "average" but really we're talking about guys who often don't get more than a cup of coffee at the MLB level unless they're either playing off the reputation of better seasons or are really good at an aspect of the game while being awful at another. These are, after all, players that most teams don't want to have on their roster if they have any decently better option.

They hypothetical replacement-level player is the skill/production level that sets where a value of 0.0 is on the scale

For context, if I look for players who played for the Blue Jays in the last 20 years who provided roughly replacement-level play over at least 200 plate appearances (not a full season's worth of play, but enough that that whatever they did can't be chalked up to small-sample randomness.)

On the position player side of things you see names like:
David Eckstein
Mark DeRosa
Eric thames
Corey Patterson
Richard Urena
Ryan Goins
J.P Arencibia

all of whom were somewhere between about negative to positive 0.2 WAR

for pitchers with a threshold of at least 100 IP (as a rule of thumb it's harder to get pitchers who put up low or negative WAR totals over long stretches because teams are more likely to replace them with better players or at least try someone else more quickly. Because a replacement level position player might do one thing exceptionally well (hit for power, field, run) that keeps them valuable, while a replacement level pitcher that can't hack it as a starter probably ends up in a bullpen in a low-innings role before they can really accumulate a ton of work)

the replacement-level all-stars for the Blue Jays are just 5 players, most of whom were largely used as relievers:

Thomas Pannone
Chad Jenkins
Bob File
Esmil Rogers
Justin Miller

Just outside my -0.2 to 0.2 cutoff was the failure of lauded prospect Kyle Drabek, who notched -0.3 WAR in 172 career Blue Jays innings across 39 games

So then, if that's what "Replacement" is, then what is the "Wins above" part?
So we've established what a replacement level player looks like. And the answer is "pretty bad". But what does that mean for the rest of the WAR stat?

research and math suggest that a hypothetical team made up of nothing but replacement-level players would be expected to win probably around 47-48 games in a 162 game season. Thus every time you replace a replacement-level guy with someone good, the team's expectations of wins increases. And if you were forced to play someone who was abjectly terrible for an extended period of time, obviously it hurts the team and they likely win fewer games.

Take, for example, one of the sets of projections for the 2021 Blue Jays. They peg Semien, Bichette, and Springer to each be worth about 4 WAR (for comparison their projected MLB leader is Trout at 6.7 with Bregman and Mookie Betts also in at 6 or better. The model, ZiPS is usually considered fairly conservative when it comes to its projections, with players often outperforming them)

That means if you took those 3 players and plunked them onto a roster of otherwise replacement-level guys, that team would gain about 12 wins and would be considered about a 60-win team.

Biggio is next on the ZiPS projection list at 3.2, followed by Vladdy at 2.6, and Jansen and Kirk at 2.0 each (I'm not sure how the model accounts for playing time, but for a thought experiment like this it's not super important). Those 4 players add almost another 10 wins to the theoretical replacement team. So put them in with Bichette, Springer, and Semien and suddenly this team that still has garbage pitching is up to 70 expected wins.

You can see how this works. The difference in WAR between a player being added to a team and the player he's replacing provides a rough estimate of how much better the team is made overall by his contributions. And you can project the expectations for a team's season to a certain degree by adding 47 wins to the combined projected season WAR for all the players on a team's roster (there's research that suggests that there is a strong relationship between WAR-projected wins using this method and a team's actual record for that season. Or at least a better relationship than most other methods of estimating projected win totals)

That said, most of the time WAR's value isn't in saying "X provides Y more wins to the team than Z" or "a collection of A, B, and C, turn this team that would win D games into one that wins E games instead") Usually it's just meant for player comparisons

alright, so why use WAR?
Like I just noted above, the chief use of WAR is as a point of player comparison over the course of a specific time period (usually a single season or their career). What the number "means" in terms of wins above replacement isn't immaterial, but is less important than understanding what the difference between compared players represents or what a player's total relative to the scale of expectations I noted way back at the beginning means.

Differences of significantly less than 1 WAR are often kind of splitting hairs. The actual on-field value of a guy at 3.3 WAR isn't really different enough from someone at 3.6 that you would want to cite those WAR values as strong evidence of anything. Rather they're an invitation to look at the rest of the player's statline, see where their value comes from and decide which player you like better in that regard. Do you prefer the guy who accumulates that value through playing good defence at a premium position? or the one that is perhaps a poor defender or plays a defensively unimportant position (DH, 1B) but puts up far superior offensive numbers? Do you want the guy that's a contact hitter who runs the bases well? Or the one that is a higher-strikeout-rate thumper who is going to hit more HRs? It depends on what you need, obviously.

But there is value in using the comparisons to see if players have more stark differences. If a guy has a 1 WAR or greater difference then it obviously represents a noticable upgrade or downgrade. Of course you can easily see that by looking at the rest of a player's stat line or potentially maybe through the eye test of their play and production, but the point is that WAR is best used as a starting point or surface level comparison to tell you if it's worth your time to look at the deeper "why" of a player's performance/value and/or compared to other players. Its also a shorthand. A means of being able to say that Guy X is way better than Guy Y without having to rattle off a half dozen stats to give the full picture.

It's value is also over time because every year's WAR is set relative to the "replacement-level" of the period. So a guy who's a 3 WAR player in 1919 is just as good (in theory) relative to his peers to a guy who's a 3 WAR player in 2019. It may not be that simple because the further we go back the fewer stats are available and the more difficult it is to trust that everything is accurate and consistent, but relatively speaking it's good enough for what we need.

If you're curious, the top 10 individual season WAR totals of all time by a hitter are dominated by Babe Ruth, who owns 6 of the 10 places on the list. The 4 others are the two most juiced up seasons of Barry Bonds' career, Lou Gehrig's crazy good 1927 MVP season, and Rogers Hornsby's 1924 season.

There's more than one version of WAR?
Yeah, this is a bit confusing. The two sites most responsible for putting WAR out there, Fangraphs (whose version is usually abbreviated as fWAR) and Baseball-Reference (either bWAR or rWAR), each calculate things slightly differently

The intent is the same, and they are moving towards trying to unify the calculation so that it is identical, but there just have been aspects that they disagree on in terms of what stats get used to form the basis of things.

On the offensive side of things Fangraphs prefers wRC+ as its base offensive value stat. Baseball-Reference uses OPS+. I believe consensus is that the differences here tend to be negligible for the most part when it comes to final total WAR calculations.

For the defensive component Fangraphs uses the defensive stat UZR (Ultimate Zone Rating) and Baseball-Reference uses DRS (Defensive Runs Saved). They're very similar stats that judge how well a player at a position gets to balls and makes plays on them (or fails to do so) relative to their peers. But they aren't identical. I don't want to get into the specifics though because I'm not 100% on them and it's kind of outside the scope of this long post.

There's a difference in the baserunning component as well, but that's not really something that most people look at the stats for so it's harder to understand and baserunning value is probably the least important aspect of WAR relative to fielding and hitting/pitching.

For Pitchers Fangraphs uses FIP while Baseball-Reference uses Runs Allowed per 9 innings. FIP isolates pitcher valuation to just the things they can control (strikeouts, walks, HRs, infield fly balls where there's no threat of it carrying over the fence from park to park) where as runs allowed includes some granting of impact to the balls that make it into play.

Which one you want to use mostly depends on what site you're more likely to frequent. I use Fangraphs more than Baseball-Reference so I tend towards fWAR. I think the consensus is that for hitters the difference between the two versions is negligible for the most part. Just for example I looked at George Springer's career. In any given season his fWAR and bWAR are usually within 0.1-0.3 of each other. His 2019 bWAR was 6.4, which is basically the same as his fWAR which I already stated earlier of 6.5. Only 2017 had a bigger difference and that was still just 0.5 (5.0 by bWAR, 4.5 by fWAR)

For pitchers I think people tend to like bWAR because it factors in better for pitchers that do well to limit hard contact or succeed with balls in play (ie groundball heavy pitchers) whereas fWAR doesn't like those because balls in play require the defence to make plays and that's not something the pitcher controls. Again the biggest thing is clarity of which you're using and consistency in using the same thing over different examples or multiple points of comparison for the same player or group of players.

But don't those differences and vagaries make it unreliable or imprecise compared to other stats?

As I said earlier, its chief value is as a sort of overview stat. It's a quick-and-dirty analysis tool, a conversation starter, a scope-narrower, or a point-of-comparison basis. It's not meant to be an argument ender unless it points to an obvious conclusion. Someone arguing that player X is better than player Y can easily have their point bolstered or refuted if someone mentions that X's WAR is 3.0 points higher/lower than Y's. Or you can say that Z from 1980 had a similarly impactful season to A from 2015 and start out by showing that their WAR totals were within 0.3 of each other before you break down the guts of comparison.

That's its use. As a way to stop people from having to trot out "well he had X, Y, Z, A, B, C for average, OBP, HRs, defensive impact, etc" every time you want to talk about a player's overall impact. Becuase 1 number provides a more focused piece of evidence or example than making people wade through a half dozen numbers and contextualize all of them.

The misuse generally comes from trying to use it in close races. Saying that X had 3.5 WAR and Y had 3.7 WAR last year is meaningless because it's so similar. It's like trying to argue that one guy was better than the other because he hit .315 over .303 or that a pitcher was superior because he struck out 21% of batters he faces whereas another "only" struck out 19%. Those arguments need more analysis because you have to differentiate the players and break the ties/close races. and WAR isn't designed to do that.

Also there's the common complaint about how different places use different versions of the stat. And yeah it makes things confusing because you might sometimes have to ask people which version they're using. But it doesn't make it "wrong" as long as you're using them in a consistent manner.

If you asked 5 bakeries to make you a chocolate cake there's no guarantee that you would get 5 identical cakes. You probably wouldn't. But that doesn't make any one of them be less of a cake than the rest. I mean unless they show up with a box that contains an apple pie with a Hershey bar sitting on top of it. WAR is a category of stat, akin to "chocolate cake" being a category of baked goods with room for variance. Whereas something like ERA is "toll house cookies" or OPS is "traditional black forest cake" and are both specific recipes with particular ingredients and steps (which doesn't stop people from modifying them, but at that point they become modifications and not "true" representations of that good)

and I think I need to stop with the baked goods metaphor before I get hungry :laugh:


I know you wanted simple, but I went for a basic overview and then more detail if you were looking for an explanation of some of the underlying concepts. But without actually getting into the messy math of it (because I don't think it's necessary. I always point to how many old school footbal people use Passer Rating or ESPN's new QBR when judging NFL QB play and have done so for decades even though the way it's calculated is an ungodly complicated mess. I don't need to see the sausage made as long as I trust that it's made well and serves its purpose in the end.

I also welcome anyone correcting me on the above because I know I probably made some mistakes or bad assumptions. It was bound to happen since I wrote this over multiple pockets of time throughout the day with long breaks in between and my train of thought may have wandered.
 
Scott Mitchell was just on overdrive and said “unless Kirk forgets to catch the ball, he is making the team”. He said the Jays will likely try to trade McGuire and probably get a player to be named later. Then look for a 3rd catcher to be a vet on the taxi squad.
 
Hope Stripling has a good outing. I want him to rebound to his pre-2020 form and be a good back-end starter.

Scott Mitchell was just on overdrive and said “unless Kirk forgets to catch the ball, he is making the team”. He said the Jays will likely try to trade McGuire and probably get a player to be named later. Then look for a 3rd catcher to be a vet on the taxi squad.

Makes the most sense. The chances of Mcguire recouping enough value as a backup to be worth holding onto and keeping Kirk down are extremely low.
 
i
 
Status
Not open for further replies.

Ad

Upcoming events

Ad