Could a consistant 5th starter put us in playoff contention?

[Michael Bradley]

Like Schad said, it's pretty ridiculous to expect a six win swing from your worst starter, let alone your ace, which is why I find it ridiculous when MB said that Roy Halladay made a 10 win difference by himself. Baseball just doesn't work like that.

It does when the disparity between top and bottom is as great as it was when Halladay was here (mostly the early years). The Jays with Halladay were typically top heavy. In 2002 for example, Halladay had a 2.93 ERA (6.9 WAR) in 239 innings. The rest of the starters (Carpenter, Loaiza, Parris, Walker, and Miller) had a collective ERA of 5.41 in 507 innings. The best of that group was Pete Walker at 4.79 in 115 innings. Yet they finished a couple of games off from .500. Replace Halladay with a league average starter and that team is horrendously bad. The pitching continued to stink after Roy for years until about 2006 or so. A team without a capable 2nd guy in the rotation to win 80-ish games for a few years without a top level offense seems pretty obviously to be on back of Halladay.

In 2005 Doc took that Kevin Mench line drive off his leg and ended his season on July 8. The Jays were 44-42 at the point. They finished 80-82. He was essentially replaced by Downs and McGowan who combined for a 4.98 ERA. So even if you think the entire team record was decided by that one change alone, the difference between an ace who got off to a 12-4 start with a 2.41 ERA and a 4.98 ERA replacement, which would be a #5 starter, was only 4 wins.

You conveniently ignored that to get to 44-42 in the first place they needed Halladay to pitch like he did (he was on his way to his best season ever). You think half a season of playing sub-.500 ball after Roy's injury means the Jays would have been just a few wins short of 80 if Halladay was replaced by Scott Downs for the whole year? Wow.

So in 2002, replace Halladay with a league average starter, and the team lead by Pete Walker in the rotation is going to win a few games less than the 78 they ended with? Ok. You believe that.

Look at St. Louis this year. Albert Pujols, I think we'd agree, is better than Halladay. His team is 80-75 right now. By your logic, if he was replaced with a league avg hitter, they would have like 68 wins. It's just not like basketball where you take away an elite player and the team is totally crap.

So according to you I created a 10 win criteria (which I don't remember doing unless it was to illustrate that Roy carried 75-win talent to 85 win seasons), which now increases to 12?

Remove Pujols from the team and replace him with an average hitter (assuming everything else stays constant) and the team is a lot worse. 5, 10, whatever. Worse. Substantially worse. He is the best hitter in the game. The downgrade from Pujols to league average is astronomical. For anyone to suggest it would only be a 2 or 3 game loss if a 1.000 OPS guy is replaced by an .750 OPS guy (or whatever the league average is), that is pretty naive.

On the flip side of that, the Jays had a ton of 5+ ERA starters in 2003, Escobar at 4.29, and Doc. They still won 86 games. If you replaced Doc with a league average starter, would that mean they would've won only 60-something or even 70-something? Of course not. They would very likely still finished with 80+ wins.

So now it goes from 10 games to 12 games to 20 games worse? Where are you getting these arbitrary numbers from? If you are using something I said as hyperbole to illustrate a point, I don't see what that is supposed to prove.

You constantly made the argument in the past that the Jays pythag showed they were better than they were, and my argument was Halladay weighed the pythag in Toronto's favor more than it should have been (for example, they were +2 without him in 2003 and +66 when he started). Yet you would point to the run differential as if Ricciardi built a great team around Roy that was stunted only by the division, which is flat out incorrect. Halladay alone carried that group to a much better mark than they had any right to be (except in 2006-08 where the talent base got better).

The reason the team is better this year has nothing to do with losing Roy; it is improvement from damn near everyone else. Put Roy on this team over Tallet/Litsch/etc and it is better. How much better? Who knows, but better. Less strain on the pen, more CG's, etc, etc.

The difference between league average and slightly better isn't much. But if you think the difference between league average and elite is this small, then I don't know what more I can say.

[Hendrix]

I did the pythagorean w/l for our team if you suplant Halladay for our #5 crop, and it comes to 91.57 wins if you use FIP, and 89.77 wins if you use ERA. The Pyth for our current team is 82.25 wins.

So from what I see Halladay would make our current team 7.75 wins better then it currently is if he took the place of our #5 starting batch. And I believe that doesn't include the additional impact we'de see from not having to go to the pen as much.

[Hoopstarr]

You conveniently ignored that to get to 44-42 in the first place they needed Halladay to pitch like he did (he was on his way to his best season ever). You think half a season of playing sub-.500 ball after Roy's injury means the Jays would have been just a few wins short of 80 if Halladay was replaced by Scott Downs for the whole year? Wow.

So in 2002, replace Halladay with a league average starter, and the team lead by Pete Walker in the rotation is going to win a few games less than the 78 they ended with? Ok. You believe that.

Yea actually, I do believe that. A few, as in 6-8 max, which happens to be around what his WAR was at the time. So they would've been something like 82-84 wins if Doc pitched the rest of 2004. That's a lot more logical than your 86+. Joe Posnanski did a piece on WAR recently. He used the following example to illustrate why WAR-defined impact makes sense:

"The question I asked a few hundred words back was this: How many more wins would this year's Pittsburgh Pirates win if they had 1923 Babe Ruth in right field instead of Lastings Milledge? It's a make-believe question without a real answer. Maybe the Pirates would pitch Ruth on days he didn't play right. Maybe teams would intentionally walk Ruth every single time (which would actually ADD to his value). Maybe other players would be inspired by Ruth. Maybe they would play even worse because he was around. Maybe none of that would happen.

At the moment, the Pirates are on pace to win 55 games, lose 107. WAR says Ruth in 1923 had a WAR of 14.7. WAR says that Milledge, in 2010, is actually worse than replacement by 1.1 wins. So adding the numbers together, WAR makes the claim that 1923 Ruth would make the Pirates 16 wins better -- taking them from 55-107 to 71-91. That seems to me about as good an answer as we can find right now."

So the best WAR season ever, 14.7, would make for a 16 game difference. You seem to think Doc was having all time great season after all time great season back then.

So according to you I created a 10 win criteria (which I don't remember doing unless it was to illustrate that Roy carried 75-win talent to 85 win seasons), which now increases to 12?

Remove Pujols from the team and replace him with an average hitter (assuming everything else stays constant) and the team is a lot worse. 5, 10, whatever. Worse. Substantially worse. He is the best hitter in the game. The downgrade from Pujols to league average is astronomical. For anyone to suggest it would only be a 2 or 3 game loss if a 1.000 OPS guy is replaced by an .750 OPS guy (or whatever the league average is), that is pretty naive.

Well, I did get them from you. You brought up the 10 game impact, so Pujols would presumably make a 12 game impact because he's better than Doc. The 20 games is just an assumption based on the exponential impact of the starters being awful from 1-5 without Doc, and that too only if you're ignoring the offensive side as I said in the last post.

I think this is more of a semantics issues than anything. You think "substantially worse" would be only 5 games. I think substantially worse is more like 10. A 5 game difference is nothing. A record could swing by 5 games entirely on run differential luck alone. So the points we're arguing on are not in alignment.

You constantly made the argument in the past that the Jays pythag showed they were better than they were, and my argument was Halladay weighed the pythag in Toronto's favor more than it should have been (for example, they were +2 without him in 2003 and +66 when he started). Yet you would point to the run differential as if Ricciardi built a great team around Roy that was stunted only by the division, which is flat out incorrect. Halladay alone carried that group to a much better mark than they had any right to be (except in 2006-08 where the talent base got better).

Oh God, we're doing this again. I will just point to this and leave it at that - viewtopic.php?p=22771519#p22771519

The reason the team is better this year has nothing to do with losing Roy; it is improvement from damn near everyone else. Put Roy on this team over Tallet/Litsch/etc and it is better. How much better? Who knows, but better. Less strain on the pen, more CG's, etc, etc.

The difference between league average and slightly better isn't much. But if you think the difference between league average and elite is this small, then I don't know what more I can say.

It's not small. I'm saying it's not the one-to-one relationship you make it to be. There's a self-correcting mechanism that makes up for losses, gains, and luck a team has from year to year and even within a year. I can't explain it numerically, but it's just regression and luck. You think this year's team is an improvement across the board? This year's offense got much worse but the pitching got much better. Even with the crazy SLG%, the runs are down from 4.93 to 4.61. The ERA improved from 4.47 to 4.26. This even trade is reflected in the team WARs the last two years: 39 and 39.3 respectively. Nearly identical. One with Doc, one without. It's further supported by the Pythag records: 84 wins last year, 82 so far this year.

[Michael Bradley]

You conveniently ignored that to get to 44-42 in the first place they needed Halladay to pitch like he did (he was on his way to his best season ever). You think half a season of playing sub-.500 ball after Roy's injury means the Jays would have been just a few wins short of 80 if Halladay was replaced by Scott Downs for the whole year? Wow.

So in 2002, replace Halladay with a league average starter, and the team lead by Pete Walker in the rotation is going to win a few games less than the 78 they ended with? Ok. You believe that.

Yea actually, I do believe that. A few, as in 6-8 max, which happens to be around what his WAR was at the time. So they would've been something like 82-84 wins if Doc pitched the rest of 2004. That's a lot more logical than your 86+.

LOL. Then "my" 86+? Care to explain where I said that?

Halladay had a 5.5 WAR in 140 innings in 2005. That was only for half the freakin' season, yet you seem to suggest the 44-42 mark prior to his injury with him was something easily to duplicate even if he wasn't there. That's a load of crap. And I never said a win total. You did.

Well, I did get them from you. You brought up the 10 game impact, so Pujols would presumably make a 12 game impact because he's better than Doc. The 20 games is just an assumption based on the exponential impact of the starters being awful from 1-5 without Doc, and that too only if you're ignoring the offensive side as I said in the last post.

I think this is more of a semantics issues than anything. You think "substantially worse" would be only 5 games. I think substantially worse is more like 10. A 5 game difference is nothing. A record could swing by 5 games entirely on run differential luck alone. So the points we're arguing on are not in alignment.

Again, bringing up 10 wins was not meant to be something statistical. If you think I passed that number off as fact, then you are wrong. I just said his impact was far greater because he carried a bunch of sub-.500 talent to +.500 records when the team had no business doing it. And I like how you increase the win total to 12 and 20 and then pass it off as if that is something I believe. Thanks for speaking on my behalf. Personally, I don't even look at stats as anally as some of you do. Take out an elite player and replace him with inferior talent and the team is worse. The team benefitted from Halladay a lot more pre-2010 than you want to give him credit for.

I am not even arguing semantics since you brought me into this thread for no reason, which is the only reason I am even posting in this topic. I couldn't care less about win difference to be honest.

Oh God, we're doing this again. I will just point to this and leave it at that - viewtopic.php?p=22771519#p22771519

What happened to run differential? The Jays scored X amount of runs more than they gave up, therefore they were unlucky in their win total, correct? Wasn't that your argument in the past? Ultimately, yes, I don't want to go through that again. Ricciardi was just an unlucky man. The team couldn't win without Roy and their +/- was a hell of a lot worse....but yes, no difference at all.

It's not small. I'm saying it's not the one-to-one relationship you make it to be. There's a self-correcting mechanism that makes up for losses, gains, and luck a team has from year to year and even within a year. I can't explain it numerically, but it's just regression and luck. You think this year's team is an improvement across the board? This year's offense got much worse but the pitching got much better. Even with the crazy SLG%, the runs are down from 4.93 to 4.61. The ERA improved from 4.47 to 4.26. This even trade is reflected in the team WARs the last two years: 39 and 39.3 respectively. Nearly identical. One with Doc, one without. It's further supported by the Pythag records: 84 wins last year, 82 so far this year.

The "improvement from everyone else" line was intended to focus on the starters in particular. We basically had four #2 starters in the rotation rather than Doc and a bunch of crap (Tallet, a rookie Cecil/Romero, Purcey, Richmond, etc). Of course the team was going to look better as the talent was spread out rather than focused on one spot. Bautista busting out and Wells getting back on track also helped greatly. The team is better now than it was last season. Nothing to do with losing Doc.

I think you are trying to pass it off as me thinking Halladay is the equivalent of Kobe Bryant in basketball. I never once said a star in baseball is worth as much as it does in basketball. That was something you want to put in my words. That is up to you. You bringing me up here is the equivalent of me saying "Hoopstar thought JP Ricciardi was the greatest GM past, present, and future" in a future thread, which I may do just for my own amusement. Hyperbole can be obnoxious when used in the wrong context.

[Lateral Quicks]

All else remaining equal (bullpen, offense, defense, other four starters) the statistics do seem to bear that out. I wouldn't go so far as to say they'd make the playoffs though. They would be in contention with 90 or so wins, but that probably isn't good enough for the playoffs.

That said, I expect the Rays and Yankees to be worse next year, at least by a couple wins apiece. Although with the Yankees, you just never know how many all-stars they'll buy to bolster their ranks in the off-season. There could be an opportunity to sneak in to the playoffs next year or the following year.

On another note, how about Vernon Wells. Maybe it was the just the injury that was holding him back. I was not expecting 31 HR, 85 RBI and an .850 OPS from him this year. A pleasant surprise to be sure.

[Hoopstarr]

"Halladay alone was turning 70+ win teams into .500 or better teams."

That's a direct quote from you in August. We can keep going back and forth if you want but what's the point if you refuse to accept what you said.

[Hendrix]

kinda bored so i tried to see what a replacment player would look like in halladay's shoes and if the difference in WAR would equal the difference in PYTH w-l. If anybody sees anything that is wrong let me know.,

For the 2002 season I did the Pyth for a team with halladay, and a team that was replaced with a pitcher that produced 0 WAR. It seemed from looking around that a player with 0 WAR was ~6.10 FIP, so I used an ERA of 6.10 in Halladay's spot. I also assumed that the pitcher would only go 5 innings opposed to halladays 7 on average. So I bumped up the pen's innings and used their ERA from the innings they already pitched as one of the numbers in calculating what our entire pitching staffs ERA was, and thus how many runs we would have given up. I think it could be argued that the pen could have had a worse ERA over this period though as their IP up to 569.


So our runs scored is 813 for the season. We gave up 828 with halladay in the equation, and 911.67 with the replacment in there. Works out to a Pyth w-l of 80-82 with him, and 72.54 without him. The relievers portion of the ERA for all this is 4.43. If you do assume that the relievers would perform worse with the higher load and input 4.70 ERA (no idea if this is an appropriate amount) for them you end up with a Pyth w-l of 71.16 wins.

Halladay had a WAR of 7.8 for that year. Which seems to be reasonable.

[Hoopstarr]

A 6.10 FIP for a replacement level starter doesn't sound right at all. David Bush and Rodrigo Lopez had the worst FIPs this year at 5.20. I like your method, but what would it be if you used a 5.20 ERA?

Edit: Probably better to use 2002 numbers. The worst FIP was Mike Hampton's 5.53 and the next three were around 5.00, so 5.20 makes sense either way.

[Hendrix]

I'm not entirely sure how to figure out what a 0 WAR player was so I looked around, and that seemed reasonable. For example one of the players was Danny Stark in 2002, and he had an FIP of 6.16 and a 0.1 WAR over 116 innings.

http://www.fangraphs.com/winss.aspx?tea ... 02&month=0

5.20 fip seems very low for the highest possible FIP to me. For example Justin Miller, and Chan Ho Park accumulated some some WAR points with FIP's in the 5.05 range in 2002.

But if you were to plug a 5.20 in . With the compensation of 4.70era for relievers you end up with 72.5 wins. And with the actual era of 4.43era for the relievers you get 73.94 wins.

[Modern_epic]

A 6.10 FIP for a replacement level starter doesn't sound right at all. David Bush and Rodrigo Lopez had the worst FIPs this year at 5.20. I like your method, but what would it be if you used a 5.20 ERA?

Edit: Probably better to use 2002 numbers. The worst FIP was Mike Hampton's 5.53 and the next three were around 5.00, so 5.20 makes sense either way.

Is that among starters by some definition, or all pitchers? Anyway, it certainly has some inning qualification, which would lead to a bias in your sample. Genuine replacement level guys are never going to get enough innings to make it through the season as a starter. A full season of replacement level starters would probably run you through 4 or 5 different guys in the role, some performing mildly above and some below replacement.

[Hoopstarr]

I'm not entirely sure how to figure out what a 0 WAR player was so I looked around, and that seemed reasonable. For example one of the players was Danny Stark in 2002, and he had an FIP of 6.16 and a 0.1 WAR over 116 innings.

http://www.fangraphs.com/winss.aspx?tea ... 02&month=0

5.20 fip seems very low for the highest possible FIP to me. For example Justin Miller, and Chan Ho Park accumulated some some WAR points with FIP's in the 5.05 range in 2002.

But if you were to plug a 5.20 in . With the compensation of 4.70era for relievers you end up with 72.5 wins. And with the actual era of 4.43era for the relievers you get 73.94 wins.

But I thought the idea was to replace Halladay with a league average starter as the ace, not the worst starter in the league. I did make a mistake in looking at qualified starters only. However, even if you go to 70+ IP starters, only two would be worse than 6.10 in 2002 so 5.20 wouldn't be average. A league average AL pitcher in 2002 was at 4.46 according to B-R (4.57 for starters, 4.25 for relievers). Then again, since finding a league avg replacement isn't exactly easy, using 5.20 is more realistic than 4.46. Long story short, your 73.94 wins sounds pretty logical to me.

[Hendrix]

[But I thought the idea was to replace Halladay with a league average starter as the ace, not the worst starter in the league. I did make a mistake in looking at qualified starters only. However, even if you go to 70+ IP starters, only two would be worse than 6.10 in 2002 so 5.20 wouldn't be average. A league average pitcher in 2002 was at 4.46 according to B-R, but since finding a league avg replacement isn't exactly easy, using 5.20 makes more sense. Thus your 73.94 wins sounds pretty logical to me.

We could do both. I was more interested in seeing what the disparity between a 0WAR pitcher (A replacment pitcher by WAR definition), and a 7.7 WAR pitcher showed in the pyth w-l to see if it equaled out.

But I could plug the league average into the equation. According to baseballreference the league average for 2002 was 4.46.

It spits out a 73.68 wins if the bullpen has 4.7ERA and the guy taking halladays spot has a 4.46 era

and 75.1 wins if the bullpen has a 4.43 era and the guy taking halladays spot has a 4.46era

[DonYon]

guys... let's just remember that all this talk is purely in hindsight. Halladay might have made majority of starts in the games where we scored no runs, or been injured, or heck even had a bad year had he played this season with us. Some even say this guy is cursed lol

Obvious stuff, but keepin' it real just in case this get too heated lol