Post by Ben (Rays GM) on Jul 19, 2015 0:28:36 GMT -5
So, with 20 votes cast, the vote currently stands at 19-1 in favor of expanding to a 6-team playoff. We will expand to a 6-team playoff starting with the 2016 season.
In this thread I'd like to give you a chance to suggest and weigh pros and cons of different options for the format of the 6-team playoffs. It would certainly require the first week to be a "bye" week for the 1 and 2 seed, while the 3 seed faces the 6 seed and the 4 seed faces the 5 seed.
How do we want to do seeding? Do we want to give the division winners the highest seeds, or do we want wild card winners to have a chance to earn these seeds if their records are better than those of some of the division winners?
Also, do we want to allow re-seeding after each round? In other words, in a typical 6 team bracket the winner of the 4 vs 5 matchup plays the 1 seed, but we can change it so that if the 6 team defeats the 3 team then the 6 team would be the one facing the 1 seed since they have the lower seeding.
Last Edit: Jul 19, 2015 0:31:14 GMT -5 by Ben (Rays GM)
Post by Ben (Rays GM) on Jul 19, 2015 16:32:38 GMT -5
My preference would be to award the division winners the top three seeds. However, the problem that this could present is the way a 6 team playoff is typically seeded:
1 vs Bye-- } 1 vs winner (4 vs 5) 4 vs 5-----
2 vs Bye-- } 2 vs winner (3 vs 6) 3 vs 6-----
Ideally, in my opinion, this would be flipped, and the 1 seed would play the winner of the 3 vs 6 seed. This way the top team would either get the worst of the division winners (3 seed) or the worst of the wild card winners (6 seed). Unfortunately there's no way to change it, but we could re-seed after each round; that way the 1 seed would get the lowest remaining seed (so if #6 upsets #3, then #1 would face #6, but if #3 defeats #6, then #1 gets #4 or #5. This still leaves it so that the most likely scenario is that the 1 seed ends up facing the 4 seed, however, which could very well be the second best team in the league (the top Wild Card winner could be as high as the second best record). Similarly, if the 3rd division winner is particularly bad (as has happened a few times already), it actually incentivizes being #6 seed - which means the WC winner with the worst record is actually in the best position.
With all that in mind, I think it might be better not to give the top seeds to the division winners. If a Wild Card winner has a better record than the second best division winner, let that WC have the #2 seed. Then we can still re-seed after each round if we want, so no matter what, best record plays worst record. This way there's never any incentive to lose a matchup in the final weeks in order to get yourself a more favorable playoff seed.
Post by Ben (Rays GM) on Jul 20, 2015 22:19:25 GMT -5
Anybody out there not okay with the idea of a wild card team potentially grabbing the number two seed and getting a bye in the first round of the playoffs?
Post by Micah (White Sox GM) on Jul 22, 2015 14:27:19 GMT -5
I think I would rather not have the wild card get the first round bye. You need to take into account the level of each division, since teams are playing extra games against the teams in their division. Allowing wild cards to be seeded higher than division winners would really make having divisions practically pointless.
Take into account the standings so far this year:
Blue Jays are 12-2 and hold the wild card in the AL. Tigers are 11-3 and hold the AL central lead.
But Blue Jays strength of schedule is at 133-164 or .448 winning pct, while Tigers strength of schedule is 139-160 or .465, because of the stronger schedule of the AL central teams. Do we want the Blue Jays at 12-2 with .448 strength of schedule seeded higher than Tigers at 11-3 and .465?
Same situation the NL: Rockies 13-1 wild card leader; 139-157 .470 strength of schedule Mets 10-3 division leader; 147-151 .493 strength of schedule
I think I would rather not have the wild card get the first round bye. You need to take into account the level of each division, since teams are playing extra games against the teams in their division. Allowing wild cards to be seeded higher than division winners would really make having divisions practically pointless.
Take into account the standings so far this year:
Blue Jays are 12-2 and hold the wild card in the AL. Tigers are 11-3 and hold the AL central lead.
But Blue Jays strength of schedule is at 133-164 or .448 winning pct, while Tigers strength of schedule is 139-160 or .465, because of the stronger schedule of the AL central teams. Do we want the Blue Jays at 12-2 with .448 strength of schedule seeded higher than Tigers at 11-3 and .465?
Same situation the NL: Rockies 13-1 wild card leader; 139-157 .470 strength of schedule Mets 10-3 division leader; 147-151 .493 strength of schedule
I've been abstaining from this discussion because I do feel I'm biased by my position in the standings, but I don't feel the stats you presented back up your point.
You talked about accounting for "level of each division" when referring to whether Max or I should be the 2nd seed, but you looked at strength of schedule at this point in the season, which isn't fully indicative of that since most teams haven't played their 2nd matchups against their divisional opponents yet. If you look at the actual "level of each division," the AL East teams have a combined record of 43-26-1 while the AL Central teams have a combined record of 38-31-1, 5 games worse (or if you factor Max and I out and just look at our divisional opponents, my divisional opponents' records would be 4 games above his). My strength of schedule may be slightly less difficult than Max's at the moment, but that doesn't reflect my full schedule with me having to play against a stronger division a 2nd time.
Post by Ben (Rays GM) on Jul 22, 2015 14:54:48 GMT -5
Interesting take, Micah. As a counter to that though, if the Wild Card has the second best record and has the chance to earn the #2 seed, that means that the #1 seed has to be from their division, which means they have to play the top team twice, which other teams might only have to play them once. The Rockies are 13-1, and that one loss was against the #1 team and their division rival, the Giants. The Mets tied the Giants, but they also lost to the Reds and Dodgers (both teams that aren't in the Mets' division, and both teams that the Rockies beat), and the Rockies themselves. When we're talking about teams with as few losses as the top seeds tend to have, a potential extra game against an elite division rival is a lot more significant than a couple percentage points in overall opponent record.
It's still a good point, and like I said in an ideal world I'd much rather that the division winners get the top seeds. I just think incentivizing losses is a far more significant problem, which the other way definitely does.
Last Edit: Jul 22, 2015 14:55:36 GMT -5 by Ben (Rays GM)
Post by Chris (Former Cubs GM) on Jul 22, 2015 15:09:36 GMT -5
I think the division winners should be the top 3 seeds with the 2 best records getting byes. The remaining 3 teams get seeded by record. I don't mind reseeding after the first round. Realistically I'm just happy that we're adding teams to playoffs and could care less how it's formatted but it makes sense to have the division winners as the top 3 seeds.
Post by Micah (White Sox GM) on Jul 22, 2015 15:33:39 GMT -5
First of all, you shouldn't feel that there's anything to be biased about or that I am picking on you specifically. This doesn't matter for this year and your team is only serving as an example because of this year's standings.
Second, I cacluated strength of schedule based on if your opponents maintain their current records throughout the season. So I did count the Rays 14-0 record against you twice even though you have only played them once at this point. So I still stand by my strength of schedule calculation.
However, you have a point that in this case, the strength of schedule doesn't necessarily stem from playing within the division. The Tigers higher strength of schedule, in this case, comes from the fact that they also play the Rays twice even though they are not in the same division.
I still maintain that allowing the wild card to seed higher than the division winners would make having the divisions pointless though. So if we seed based on record only, let's just forget the divisions with the top 6 teams from each league getting into the playoffs with no "division winner" and "wild card" labels necessary.
Post by Ben (Rays GM) on Jul 22, 2015 15:45:25 GMT -5
Three times already we've had a division winner make the playoffs with a .500 record or less. If we seed the wild card winners 4th, 5th, and 6th, then the 6th seed plays the 3rd best division winner while the other two wild cards (4th and 5th seeds) play each other. It's very easy to envision a scenario where teams would rather be the 6th seed, playing the weakest of the division winners, than the 5th seed, playing the potentially very strong top wild card team.
Here's a hypothetical scenario for this year - we're heading into the final week of the season. The Tigers have won the central, the Blue Jays have clinched the first wild card spot, and the White Sox and the Indians are going to get the other two wild card spots, but they currently have identical records and we don't know who will end up as the 5th seed and who will end up as the 6th seed. With the schedule changes I made prior to the season, the White Sox and Indians are playing against each other in the final week. The winner gets the #5 seed and has to face the #4 seeded Blue Jays in the first round of the playoffs while the loser gets the #6 seed and faces the #3 seeded Angels in the first round. No disrespect to the Angels, but it's very clear that in that scenario losing would be the more desirable outcome. The strategic move there would be to try to lose, or at the very least, not bother making an extra effort to win it. To me that's a much bigger issue than the strength of schedule thing.
Last Edit: Jul 22, 2015 15:50:32 GMT -5 by Ben (Rays GM)
Three times already we've had a division winner make the playoffs with a .500 record or less. If we seed the wild card winners 4th, 5th, and 6th, then the 6th seed plays the 3rd best division winner while the other two wild cards (4th and 5th seeds) play each other. It's very easy to envision a scenario where teams would rather be the 6th seed, playing the weakest of the division winners, than the 5th seed, playing the potentially very strong top wild card team.
Here's a hypothetical scenario for this year - we're heading into the final week of the season. The Tigers have won the central, the Blue Jays have clinched the first wild card spot, and the White Sox and the Indians are going to get the other two wild card spots, but they currently have identical records and we don't know who will end up as the 5th seed and who will end up as the 6th seed. With the schedule changes I made prior to the season, the White Sox and Indians are playing against each other in the final week. The winner gets the #5 seed and has to face the #4 seeded Blue Jays in the first round of the playoffs while the loser gets the #6 seed and faces the #3 seeded Angels in the first round. No disrespect to the Angels, but it's very clear that in that scenario losing would be the more desirable outcome. The strategic move there would be to try to lose, or at the very least, not bother making an extra effort to win it. To me that's a much bigger issue than the strength of schedule thing.
Meanwhile, let's say the Tigers and I have very similar records at the time as well. One of us will get the #1 seed while the other will get the #2 seed. Both of us would rather have the #2 seed and get to play the winner of 3 vs 6 (the Angels vs the "lesser" wild card) rather than have to play the winner of 4 vs 5 (the two "better" wild cards). So now I'm considering throwing my matchup against the Blue Jays, and the Tigers are considering throwing their matchup against the Royals - who themselves could be in the picture for that 6th wild card spot. It all gets very messy.
Post by Nick (Orioles GM) on Jul 22, 2015 16:02:48 GMT -5
I generally would have favored the division winners getting the better seeds but I think Ben's argument compelling so I would lean toward doing it by overall record.
Post by Ben (Rays GM) on Jul 22, 2015 16:42:31 GMT -5
I really just don't think giving the division winners the top 3 seeds is a viable option given the problem outlined above. I suppose just ignoring division winners entirely is something we could consider, but I don't feel compelled by that option - better to try to keep some semblance of realism, in my opinion.
Post by Chris (Former Cubs GM) on Jul 22, 2015 16:59:01 GMT -5
I know a few people wanted to move to 6 teams because realism wasn't a good enough reason to keep playoffs at 4 teams but abondoning divisions seems a bit much. I understand all the hypotheticals of the situation with division winners and it doesn't seem right that blue jays would end up a 4 seed playing the 1 seed in the ALDS but nothing is going to be perfect. I missed playoffs last year with the 4th best record in the league and Dodgers will miss this year with a solid record. It's never going to be 100% flawless.
Post by Ben (Rays GM) on Jul 22, 2015 17:08:09 GMT -5
I agree with Chris, and I think that's well said. It can't be perfect. We just need to find the best balance between our various needs. Realism is one of many things we consider, and while I agree with what Micah's saying that allowing the WC to seed higher than division winners makes the divisions less important or even necessary, I think realism is still a strong enough argument to keep them in place. Personally, I'd rather keep that option (getting rid of divisions entirely) off the table for now.
