QB platoon - Picking weekly starters (1 Viewer)

[dirty_martini]

Greetings - longtime lurker, first time poster here.

I'm planning to run a platoon of QBs in a dynasty league this year. I suspect some of you here do the same and I'm hoping you'll share your knowledge. What are the best tools to use when picking a starter for the week? SOS? Projections? The hot hand over the last x weeks? Statistical analysis of some sort? A combination?

Thanks in advance!

 

[Anthony Borbely]

Chase Stuart has a great QBBC article every year which seems to be what you are looking for. This years is not posted yet, but below is last years QBBC article so you can have an idea of what it is about. This is a well written article.

Link to last years article

 

[Sweet Love]

There are a lot of QBBC guys on the board and usually a QBBC thread gets started each year where people post who they are using, WW wonders, etc. Personally, I try to be as organized as possible and prior to drafting I evaluate SOS, ADP, etc. and come up with 3-4 guys in the QB 10-15 range that I am going to target. Once the season gets going though, I am usually swamped and ride the hot hand. If you are a subscriber, there is a spreadsheet they generate each week that shows you who has/has not been hot for the past three weeks and I find it pretty helpful during the season.

 

[dirty_martini]

Chase Stuart has a great QBBC article every year which seems to be what you are looking for. This years is not posted yet, but below is last years QBBC article so you can have an idea of what it is about. This is a well written article.

Link to last years article

Yes, I did read that. But it's more of a draft guide than how to pick the QB to start any given week. The article suggests setting your lineup for the year based on the pre-season SOS and I just can't imagine that is the most effective strategy. Afterall, defenses are not always what we expect in pre-season. I could indeed use SOS to pick a weekly starter like the article used in developing draft pairs/trios. Anyone tried that and have any feedback?I could also use the weekly projections from any number of sources to pick my starter any given week. Just wondering who has tried it and what thoughts they have to share on it ...

 

[MAC_32]

Not sure how much this will help, but sometime in the next month I'll go through several of the QB's ranked between about 8 and 20 (the ones I believe in) and cross reference their schedules. Two years ago I was able to get Bulger + Warner and ride the matchups to some supreme QB production (I believe I was 3rd in the league in overall QB scoring), I only had one week that I perceived as one of them not having a great matchup and got lucky that week with Warner having a servicable game.

I don't try to go QBBC and I don't try to get a top QB, but I like to have a QBBC plan in place in the event that I do not get a top QB in the draft. Ranking pass defenses 1-32 is impossible, but I've had success identifying who will be some of the weaker ones in the past.

 

[Sigmund Bloom]

Dodds and I do weekly subscriber projections to give you two points of view - we adjust right up til kickoff for weather and late scratches that affect the offense or opposing defense - hot hand and strength of opponent definitely figures into both.

The best general advice I can give (I know its a cliche) is to not overthink it. Generally you'll want to play the QB who has been playing better football - most of the time Ive made the wrong in call in a QBBC, it's when I give too much weight to the matchup.

 

[T Bell]

Dodds and I do weekly subscriber projections to give you two points of view - we adjust right up til kickoff for weather and late scratches that affect the offense or opposing defense - hot hand and strength of opponent definitely figures into both.The best general advice I can give (I know its a cliche) is to not overthink it. Generally you'll want to play the QB who has been playing better football - most of the time Ive made the wrong in call in a QBBC, it's when I give too much weight to the matchup.

I have never done a QBBC well. The matchup problem always kills me even when I have a good group.It sounds easy enough, but a bad game, recovery from injury, etc. all tend to cloud who "the hot QB" is at any given time.

 

[massraider]

Chase Stuart has a great QBBC article every year which seems to be what you are looking for. This years is not posted yet, but below is last years QBBC article so you can have an idea of what it is about. This is a well written article.

Link to last years article

Yes, I did read that. But it's more of a draft guide than how to pick the QB to start any given week. The article suggests setting your lineup for the year based on the pre-season SOS and I just can't imagine that is the most effective strategy. Afterall, defenses are not always what we expect in pre-season. I could indeed use SOS to pick a weekly starter like the article used in developing draft pairs/trios. Anyone tried that and have any feedback?I could also use the weekly projections from any number of sources to pick my starter any given week. Just wondering who has tried it and what thoughts they have to share on it ...

Well, one thing to do is check and see which combos Chase came up with, and how they would have scored using your scoring system. go week-by-week, pick which QB you would have started, and see how you would have done for the season.The biggest thing:

Coming up with a QBBC combo that you like isn't too difficult. You can find a few 2nd tier guys you like, go through their schedules, and come up with some really attractive matchups.

Here's my thing with QBBC (or really, any other position by committee): It's not for everyone. I know a lot of owners that have done it, and after a few weeks of picking the wrong guy, they are pulling their hair out. And the next year, they grab Peyton Manning with their 1st rounder. You need to know what kind of owner you are. If you are willing to live with choosing wrong from time to time, and don't overreact to every poor performance by a QB, it has value. A lot of owners just like knowing who they are going to start every week.

 

[TruckBoatTruck]

A shot of Jameson and a dartboard. I'm only half joking.

 

[dirty_martini]

Well, one thing to do is check and see which combos Chase came up with, and how they would have scored using your scoring system. go week-by-week, pick which QB you would have started, and see how you would have done for the season.

I did that - it was so-so. If I used the schedule Chase laid out at the beginning of the season, 50% of the time it gave me a better QB score than I got that week. That's one of the things that lead me to post this question. I like the basic idea. Particularly for this league. It's a dynasty with 10 teams and small rosters. Mostly everyone holds their stud and a young prospect. The waiver wire often has serviceable players on it. I'm trying to figure out a means to use the waiver wire as a platoon of QBs. Of course, I'll have my avg QB on the roster as a fall back.Picking the WW or roster QB with the easiest schedule that week gives me a better QB score 62 percent of the time. That's without regard for the QBs skill level and using a SOS that was calculated at the beginning of the season. If I use a Dynamic SOS and factor in QB skill I suspect I can raise that number even more.

 

[dirty_martini]

A shot of Jameson and a dartboard. I'm only half joking.

I hope the dartboard is the joking half.

 

[TruckBoatTruck]

A shot of Jameson and a dartboard. I'm only half joking.

I hope the dartboard is the joking half.

To expand a bit: I always end up drafting QBs late in redrafts, and have settled on the gut feeling method. For whatever reason, I think it's more successful. Certainly less stressful.Point being, I would never put that much stock in a dynamic SOS (or any other statistical measure, FWIW) for choosing who to start mid-season, but that's just me. For drafting, absolutely, and I give SOS some consideration for a QBBC, but it's never a primary factor - more like a tiebreaker if needed. Your idea is intriguing, though. If I were you, I'd definitely see if using dynamic SOS over pre-season SOS led to an over 62% success rate - I wouldn't automatically assume it would be better. One question - if you're going to do this, why keep that "average QB" you mentioned on your roster? Even if your success rate is over 50%, saving the roster space seems just as important as improving your overall QB scoring. Or am I missing something?

 

[dirty_martini]

A shot of Jameson and a dartboard. I'm only half joking.

I hope the dartboard is the joking half.

To expand a bit: I always end up drafting QBs late in redrafts, and have settled on the gut feeling method. For whatever reason, I think it's more successful. Certainly less stressful.Point being, I would never put that much stock in a dynamic SOS (or any other statistical measure, FWIW) for choosing who to start mid-season, but that's just me. For drafting, absolutely, and I give SOS some consideration for a QBBC, but it's never a primary factor - more like a tiebreaker if needed. Your idea is intriguing, though. If I were you, I'd definitely see if using dynamic SOS over pre-season SOS led to an over 62% success rate - I wouldn't automatically assume it would be better. One question - if you're going to do this, why keep that "average QB" you mentioned on your roster? Even if your success rate is over 50%, saving the roster space seems just as important as improving your overall QB scoring. Or am I missing something?

Three reasons for keeping that average qb - rosters are frozen during playoffs. So I'll have to carry 2 qbs then anyway. Kind of limits my ability to use that slot for stashing a developmental player. During backtesting, a lot of the players I used were backups and I don't want to be caught with a starter returning at the last minute and leaving me with no QB that week. And the third reason is to protect my self in the event I cant get a player I need off waivers.I'll let you know how this works.

 

[T Bell]

Dodds and I do weekly subscriber projections to give you two points of view - we adjust right up til kickoff for weather and late scratches that affect the offense or opposing defense - hot hand and strength of opponent definitely figures into both.The best general advice I can give (I know its a cliche) is to not overthink it. Generally you'll want to play the QB who has been playing better football - most of the time Ive made the wrong in call in a QBBC, it's when I give too much weight to the matchup.

I have never done a QBBC well. The matchup problem always kills me even when I have a good group.It sounds easy enough, but a bad game, recovery from injury, etc. all tend to cloud who "the hot QB" is at any given time.

Here's what I'm talking about. Last year in a basic redraft league (no PPR; no IDP; no bonuses) I managed to grab (in this order) Tony Romo, Donovan McNabb, and Jason Campbell as my QB's. My mindset at the time was that Romo might not be that good given his bad finish in 2006, but DMac and JC should be able to fill in when he's got bad matchups. Week 1: I opt to start McNabb @GB because of what I thought was a suspect defense instead of Romo vs. NYG. McNabb puts up 13.1 points, a number which Romo more than triples with a 39.4 pt performance. I lose by 1.1 points. Week 2: For some reason, I decide to stick with McNabb thinking that he'll put up good numbers vs. Washington while Romo will struggle @ Miami. Romo's game against the Giants was a fluke, right? The Giants just had a bad defensive game, right? Nope. McNabb throws up a 13.5, which is lower than both Campbell's 17.4, and Romo's 20.9. I lost by more than this difference though. Week 3: Ok, no more McNabb love. His knee is clearly still bothering him and he's just not up to speed. This time I'm going to be smart. I'm going to start Campbell against the Giants defense that's given up like 90 points in two games. That's what I do, and Campbell puts up a 14.7, while Romo puts up 23.5 against the Bears . . . and McNabb puts up 35.8 vs. Detroit. The rest of the way I started Romo except for his bye week . . . only to watch Campbell outscore him the next two weeks. I was otherwise fine except for one other week in which McNabb outscored Romo (week 10). Maybe I made some stupid decisions, and maybe I was just unlucky, but 13 weeks is not a lot of time for trial and error. By the time you figure out "the hot hand", you are through a good chunk of the season. It's just a very tough assignment IMHO.