Could consistency actually be a negative? (1 Viewer)

[Los Gigantes]

Have been brushing up on some great Adam Harstad articles on dynasty.  Came across this one http://subscribers.footballguys.com/apps/article.php?article=HarstadDiT3 about consistency.  I think that he understates his case.  If it is true that the majority of fantasy games are not close, an average performance does not help much.  As a thought experiment, let's assume the typical margin of a game is 20 points.  A consistent player with the same points per game (say 15) that scores 15 every week like clockwork, will rarely make a difference in the outcome.  But a player that scores 30 half the time and 0 half the time (to use an extreme example) will rarely cost you the game when he has an off day, but will win it for you when he has a good day.  Maybe "difference making potential" is far better than consistency (aka consistent mediocrity).  

 

[Anarchy99]

My goal is to roll out the higher scoring team in any given week. If there is a player that scores 0 or 30, there should a reason why his scoring is that way. I doubt the choice every week will be between the guy that scores 15 ppg every week or the one that scores 0 or 30. 

I always pick up players with higher ppg averages and more injury risk for the reason they score more when they get on the field. 

I agree that consistent scoring for non fantasy starters really isn't a big draw. So the WR45 that scores 6 points a game is better than nothing, but I generally can go match up and find a better option. 

That being said, I wouldn't suggest going for too many home run or strikeout types, especially if you don't have to.  And you should be able to have some idea based on NFL match ups what you might expect for fantasy scoring. 

 

[Hawkeye21]

When I'm picking my starters I want consistent numbers.  My bench players are the ones I'm fine with being boom or bust.

 

[Warhogs]

I don't know.  I owned Brandin Cooks in one league.  In the weeks where he was huge I blew away my opponent.  I lost a few close games when he stunk up the joint.

 

[BradtheAg]

Not in weekly H2H leagues

 

[Mr. Irrelevant]

That being said, I wouldn't suggest going for too many home run or strikeout types, especially if you don't have to.  And you should be able to have some idea based on NFL match ups what you might expect for fantasy scoring. 

Another reason it's dangerous to rely on too many boom-or-bust guys in standard H2H leagues is that players, and situations, change dramatically from one year to the next. When you draft a bunch of guys with low signal-to-noise ratios, so to speak, it takes a lot longer to determine that there's an underlying issue that might make your guy unstartable / unrosterable. And you can't get those weeks back.

Take a comparison like Vincent Jackson vs. Torrey Smith from last year - both drafted in the WR45-50 range who probably started the season in a lot of lineups. V-Jax was supposed to be the old reliable chain-moving #2 opposite Evans, but by Week 3 (4, 8, 6 points in PPR) it was easy to tell he wasn't getting the job done any more and bench / drop him. Torrey started out with weeks of 3, 14,and  6 ... but Torrey has 3-game stretches like that every season and then goes and puts up a 30-spot the next week, so too soon to cut bait, right? Nope - he averaged just 4.5 PPG the rest of the way, and cost a lot of hopeful owners who held on a couple weeks too long. Ditto similar guys like D-Jax and Fuller, though at least they gave you a couple big weeks out of the gate before they torpedoed your season.

I have no problem with high-variance guys, but as you and @Hawkeye21 referenced, I want that high variance to be either (a) due to injury risk or (b) on my bench. That way, I can ride the big weeks if/when they're hot and IR/drop them when they get banged up or fail to perform.

 

[FreeBaGeL]

 a player that scores 30 half the time and 0 half the time (to use an extreme example) will rarely cost you the game when he has an off day, but will win it for you when he has a good day.  Maybe "difference making potential" is far better than consistency (aka consistent mediocrity).  

This doesn't make sense.  A player dropping your point total by 15pts off of what a consistent guy would give you is just as likely to cause you to lose a game as a player increasing your point total by 15pts is to cause you to win it.

 

[Los Gigantes]

This doesn't make sense.  A player dropping your point total by 15pts off of what a consistent guy would give you is just as likely to cause you to lose a game as a player increasing your point total by 15pts is to cause you to win it.

not necessarily.  If without the player in question you are down by 20, 0 or 15 does not matter, you still lose.  30 you win.  Its an empirical question based on how close games tend to be.

 

[FreeBaGeL]

not necessarily.  If without the player in question you are down by 20, 0 or 15 does not matter, you still lose.  30 you win.  Its an empirical question based on how close games tend to be.

And if without the player in question you are down by 10, 15 or 30 does not matter, you still win.  0 you lose.

It's the same.

 

[lod001]

I like to have my #1 WR consistent, my #2 consistent as much as possible and my #3 like Martavis Bryant when he is playing since in a 14 team dynasty, it's tough to get 3 consistent upper tier WRs.

 

[thatguy]

I would think a guy who gets you at least 15 points every week would be just as likely to blow up for 30 as a player with higher variance.

ETA: take for instance of super reliable Jarvis Landry in PPR formats. He's pretty much a lock for 15 points per game but has had weeks where he blew up for 30 or more.

 

[thatguy]

The goal should be to load your rosters with players who consistently blow up. Easier said than done, but I think anyone who has played Dynasty for long enough has at some point had a team that was just stacked--maybe for just a season or 2, maybe for 3 to 5 or more seasons straight. And that's when it's truly fun. That is the ultimate goal. Granted it's not going to last forever but I've had some pretty amazing Dynasty teams that stayed top-heavy for anywhere from three to five year periods. 

 

[Hawkeye21]

Basically you want high floor players as your starters.  We all know a low floor player has the chance to put up a huge game but what's more important is that we know their low scoring games are still going to be better than the average player.

The players you want to stay away from are the low floor guys that also have a low ceiling.  The guys in between are the ones you stash on your bench or put in the flex.

 

[thatguy]

Pretty much sums it up. 

 

[FUBAR]

Basically you want low floor players as your starters.  We all know a low floor player has the chance to put up a huge game but what's more important is that we know their low scoring games are still going to be better than the average player.

The players you want to stay away from are the low floor guys that also have a low ceiling.  The guys in between are the ones you stash on your bench or put in the flex.

Pretty sure that's high floor.  But yeah. 

But I'm quite happy to put out consistent performers in large leagues.  If you're starting 22 players, consistency helps win every week.  

I do look for wr3/4s that can blow up.  Mike Wallace, will fuller types.  Usually they're cheap and can win a game on occasion. 

 

[Hawkeye21]

Pretty sure that's high floor.  But yeah. 

But I'm quite happy to put out consistent performers in large leagues.  If you're starting 22 players, consistency helps win every week.  

I do look for wr3/4s that can blow up.  Mike Wallace, will fuller types.  Usually they're cheap and can win a game on occasion. 

Yep.  I fixed it.  All this talk about ceilings, floors, highs and lows.  I got confused.

 

[Los Gigantes]

And if without the player in question you are down by 10, 15 or 30 does not matter, you still win.  0 you lose.

It's the same.

it's not the same.  its different.  and which is better depends completely on whether there are more close games or more lopsided games.  Without looking at the data over a number of leagues, it is ambiguous which is better.  But if there is a preponderance of higher spreads, then inconsistent is preferable if ppg is the same.  if most games are close, consistent is better.  The right answer is data driven.

 

[Arodin]

it's not the same.  its different.  and which is better depends completely on whether there are more close games or more lopsided games.  Without looking at the data over a number of leagues, it is ambiguous which is better.  But if there is a preponderance of higher spreads, then inconsistent is preferable if ppg is the same.  if most games are close, consistent is better.  The right answer is data driven.

More than this, it depends heavily on whether you are generally favored or the underdog.  If my team is strong and generally favored vs. opponents, then I want the steady 15-ppg guy.  I'm more likely to suffer a fluke loss from a 0 than I am to need a fluke 30 to pull one out.  On the other hand, if I have a weaker team, I need more of those flukes to pull off the victories, and then I'm in the "I'll lose with a 15 or a 0, so might as well play for the 30" camp.

 

[FreeBaGeL]

it's not the same.  its different.  and which is better depends completely on whether there are more close games or more lopsided games.  Without looking at the data over a number of leagues, it is ambiguous which is better.  But if there is a preponderance of higher spreads, then inconsistent is preferable if ppg is the same.  if most games are close, consistent is better.  The right answer is data driven.

Except that this whole discussion started with you saying something completely different.  You weren't claiming that there was a data driven answer to this question.  You were claiming that a guy scoring 0 would cost you a minimal number of games while a guy scoring 30 would win you a lot.  With no data.

Again, you said...

" a player that scores 30 half the time and 0 half the time (to use an extreme example) will rarely cost you the game when he has an off day, but will win it for you when he has a good day."

 

[DropKick]

Except that this whole discussion started with you saying something completely different.  You weren't claiming that there was a data driven answer to this question.  You were claiming that a guy scoring 0 would cost you a minimal number of games while a guy scoring 30 would win you a lot.  With no data.

Again, you said...

" a player that scores 30 half the time and 0 half the time (to use an extreme example) will rarely cost you the game when he has an off day, but will win it for you when he has a good day."

You are right; his proposition is flawed. 

You could talk yourself into anything with these scenarios.  Suppose a player averages 15 a week; every week.  Also, every game is decided by exactly 20 points.  I have the option (after the game is done) to substitute another player who put put either 0 or 30 points that week.  How many extra games will I win with this special privilege?  So, a 0 is 15 less points than Mr Steady scores and 30 is 15 more...  When I lose by 20, the 15 extra points isn't enough to get the win.  When I win by 20, I wouldn't have lost even if I took the zero.  So the conclusion is the 0/30 player will never impact a game. 

Taken to "an extreme" -  will you win more with eight guys that average 15 or eight guys that put of 0 or 30? 

I never think of a single player being the difference between winning and losing but it is nice when someone really goes off.  I think of it as a team and, if everyone shows up, I should be that much more difficult to beat.  I like the idea of consistency, which is not necessarily low floor.  I say "idea" because there is always some variance involved.  Do you really want a boom or bust player?

 

[Hawkeye21]

If I can have a team where each player averages 15 points a game I'd be very happy.

 

[Arodin]

If I can have a team where each player averages 15 points a game I'd be very happy.

Easily done...just play in a 4-person league with short lineups!

 

[Bronco Billy]

If the inconsistent guy scored 30 and then 0 on alternating weeks, then it might be discusion worthy. 

But when speaking to inconsistency, what enters my mind is the guy who scores that 60 FF points in that 4 week span, but he does so by getting 36 in his big week and then averaging 8 each week for the other 3.

So yeah, you win that one big week, but if your leagues are anywhere near as competitive as mine, the 7 less pts than Mr. Consistent over those other 3 weeks are more likely than not Ls.

 

[Interseptopus]

If the inconsistent guy scored 30 and then 0 on alternating weeks, then it might be discusion worthy. 

But when speaking to inconsistency, what enters my mind is the guy who scores that 60 FF points in that 4 week span, but he does so by getting 36 in his big week and then averaging 8 each week for the other 3.

So yeah, you win that one big week, but if your leagues are anywhere near as competitive as mine, the 7 less pts than Mr. Consistent over those other 3 weeks are more likely than not Ls.

Chad Johnson was like this. He'd score you 4/50 for like 4 weeks and then go off for 225/3. He'd finish the season with like 10 TDs but those 10 TDs were scored in 4 games. Meanwhile he's getting you 9 points in PPR every other week and you drafted him to be your WR2 at the very least (in his prime). So that player finishes in the top 10-15 in WRs but really he was a WR4-5 for 75% of the season and a WR1 for 25%. I suppose if you feel good about a guaranteed win 25% of the time then this is right up your alley.

Granted, Ajayi won me my superbowl this year with his inconsistency, but these types of players I usually try to avoid at all costs. 

 

[Hawkeye21]

If the inconsistent guy scored 30 and then 0 on alternating weeks, then it might be discusion worthy. 

But when speaking to inconsistency, what enters my mind is the guy who scores that 60 FF points in that 4 week span, but he does so by getting 36 in his big week and then averaging 8 each week for the other 3.

So yeah, you win that one big week, but if your leagues are anywhere near as competitive as mine, the 7 less pts than Mr. Consistent over those other 3 weeks are more likely than not Ls.

That's actually an example of a player being consistent.  I'd love having a guy on my team knowing he's going to score 30 points every other week.  I'd start him those weeks and sit him the opposite weeks. 

 

[Interseptopus]

If I can have a team where each player averages 15 points a game I'd be very happy.

Exactly. But possibly overstated. I'd even say 10-12 points 

I draft so that my WR1 and RB1 have big play potential (this year, Bell and OBJ). This year I drafted for a good QB (Brady) but I usually go QBBC. Brady fell because everyone in my league hates him. Wohoo for me. 

If my WR2 and RB2 can produce 10 points a game on average I am thrilled because that means they have potential to score 20+ a few times a season, but they are averaging at least 10 their "off" weeks. I am thrilled with this. Same with TE. If I can get 10-12 from my TE I am overly thrilled.

I will win games with my QB, WR1 and RB1. The rest of my guys just need to produce me 10-12 points. In that case I am getting 60-70 points from my "others" and looking at 17-25 from my QB, WR, RB putting me at 125+ weekly (not counting those big weeks from the "others" mind you), which should be a win in my league most weeks. This strategy works for me. I've won 4 of the last 5 seasons and took 3rd place the off year due to a very late stat correction for Denard Robinson costing me a game by a tenth of a point (would have won the superbowl that year). I'm consistently one of if not the top scoring team in the league. If it ain't broke don't fix it

 

[FreeBaGeL]

Chad Johnson was like this. He'd score you 4/50 for like 4 weeks and then go off for 225/3. He'd finish the season with like 10 TDs but those 10 TDs were scored in 4 games. Meanwhile he's getting you 9 points in PPR every other week and you drafted him to be your WR2 at the very least (in his prime). So that player finishes in the top 10-15 in WRs but really he was a WR4-5 for 75% of the season and a WR1 for 25%. I suppose if you feel good about a guaranteed win 25% of the time then this is right up your alley.

Granted, Ajayi won me my superbowl this year with his inconsistency, but these types of players I usually try to avoid at all costs. 

Man where were you all those years ago when I was arguing that Chad Johnson was overrated because of it and getting blasted for it here on these boards

The only thing I would disagree with in your post is the characterization that he was people's WR2.  He was being drafted as a top 5 WR and certainly a WR1 back then.

This actually brings up another point against the inconsistent players, which is that in practice a lot of times people miss out on the big games because they have given up on a guy and benched him after a string of bad games.

I remember that even with Chad Johnson.  He had a handful of meh games in a row then finally went off for 11-260-2 in a game, and all I remember is people complaining that he finally had a good game and he was on their bench.  And in missing that one game they missed like 20% of his fantasy points for the entire season.  Then he gets back into the lineup and for the fantasy playoffs he gave you 3-37 and 3-32  :X

And that was Chad Johnson, their first round pick.  It's a lot easier to give up on a guy when his name is Quincy Enunwa and he's had 3 bad games in a row.  It's not like we have a crystal ball that says "this guy is going to be inconsistent all season".  A lot of times it takes a while to figure that out and people are stuck thinking he's just done putting up points, and plant him firmly on their bench.

 

[Interseptopus]

Man where were you all those years ago when I was arguing that Chad Johnson was overrated because of it and getting blasted for it here on these boards

The only thing I would disagree with in your post is the characterization that he was people's WR2.  He was being drafted as a top 5 WR and certainly a WR1 back then.

This actually brings up another point against the inconsistent players, which is that in practice a lot of times people miss out on the big games because they have given up on a guy and benched him after a string of bad games.

I remember that even with Chad Johnson.  He had a handful of meh games in a row then finally went off for 11-260-2 in a game, and all I remember is people complaining that he finally had a good game and he was on their bench.  And in missing that one game they missed like 20% of his fantasy points for the entire season.  Then he gets back into the lineup and for the fantasy playoffs he gave you 3-37 and 3-32  :X

And that was Chad Johnson, their first round pick.  It's a lot easier to give up on a guy when his name is Quincy Enunwa and he's had 3 bad games in a row.  It's not like we have a crystal ball that says "this guy is going to be inconsistent all season".  A lot of times it takes a while to figure that out and people are stuck thinking he's just done putting up points, and plant him firmly on their bench.

I didn't want to say WR1 and get blasted for it, but you are right. He was my WR1 and I got burned, never did it again. I think one year he was my flex and I still hated it. I've been a big proponent of the "Chad Johnson is overrated" crowd since then. Same with many others. I got a lot of grief for this as well. He would finish as a top 12 WR every year and be drafted as such and then people would complain. I would point to his game logs and people would call me nuts. Continues to happen to this day as I identify players who perform very similarly to Chad Johnson... even in those threads I get grief for suggesting this

 

[Arodin]

I didn't want to say WR1 and get blasted for it, but you are right. He was my WR1 and I got burned, never did it again. I think one year he was my flex and I still hated it. I've been a big proponent of the "Chad Johnson is overrated" crowd since then. Same with many others. I got a lot of grief for this as well. He would finish as a top 12 WR every year and be drafted as such and then people would complain. I would point to his game logs and people would call me nuts. Continues to happen to this day as I identify players who perform very similarly to Chad Johnson... even in those threads I get grief for suggesting this

So who are some current players you have in this category? 

I've always felt Julio Jones was, but never owned him so haven't actually dug into his week-by-week numbers.  Cam is another...

 

[FUBAR]

If I can have a team where each player averages 15 points a game I'd be very happy.

Easily done...just play in a 4-person league with short lineups!

Or just adjust your scoring system. 

 

[Interseptopus]

So who are some current players you have in this category? 

I've always felt Julio Jones was, but never owned him so haven't actually dug into his week-by-week numbers.  Cam is another...

Interesting question. I don't necessarily have time to answer this in depth immediately, but I'll come back to this at a later time. Off the top of my head: 

I do feel like Julio Jones kind of fits this category but not exactly. 
OBJ was close to approaching this category this last season, but he had several games of 15 or so fantasy points without a TD that made up for it, but the first 8 games I was nervous 

Jay Ajayi is a classic example. I think he finished top 15 but had 2-3 big games. 
Diggs is a good example this year. He would either go 4/40 or 6/125/2
Cam Newton is a great example. I avoid him at all costs 


A year or two ago I had suggested Chris Johnson in his later years was similar to Chad. Early in his career he produced quite well but he kind of petered off into flirting with Chad Johnson status for a while. I got railed for suggesting this. I think that was a good debate actually. Good points on both sides

I'll add more later. I'm just commenting on players I know because I owned them. 

 

[Bronco Billy]

So who are some current players you have in this category? 

I've always felt Julio Jones was, but never owned him so haven't actually dug into his week-by-week numbers.  Cam is another...

http://www.espn.com/fantasy/football/story/_/page/consistency2016/final-2016-fantasy-football-consistency-ratings