What's new
Fantasy Football - Footballguys Forums

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

WR Chris Olave, NO (1 Viewer)

I like Olave. Good speed. Good route runner. Good athlete. The problem is he obviously has no gray area in regards to concussions. He gets a concussion he is out for year. It's like Dinero playing Russian roulette in The Deer Hunter. Just hard to predict.
 
How has he looked so far? Might have to start using him.
A lot of Targets. Not amounting to much so far though.
All you can ask for is opportunity...and a decent QB to give you those opportunities.
Well we got 1 of those. He's going to get 12 targets a game by default. The quality of those targets is up in the air.
The quality of targets hasn't hurt J. Johnson
True, but I imagine the quality of the coverage is different and Juwan has an ADOT of 7.2 and Olave is at 9.5.

Neither of those are terribly impressive ADOT but 2.3 yds is 2.3 yds.
 
How has he looked so far? Might have to start using him.
A lot of Targets. Not amounting to much so far though.
All you can ask for is opportunity...and a decent QB to give you those opportunities.
Well we got 1 of those. He's going to get 12 targets a game by default. The quality of those targets is up in the air.
The quality of targets hasn't hurt J. Johnson
I'm actually starting both this week. God help me
 
How has he looked so far? Might have to start using him.
A lot of Targets. Not amounting to much so far though.
All you can ask for is opportunity...and a decent QB to give you those opportunities.
Well we got 1 of those. He's going to get 12 targets a game by default. The quality of those targets is up in the air.
The quality of targets hasn't hurt J. Johnson
I'm actually starting both this week. God help me
I think Johnson has a good chance to keep up his performance. Seattle has a good defense for sure, but the Pittsburg TEs were able to get some play on them and there’s some dink n dunk vulnerabilities.

Olave should see his targets but I’d be more reserved in your expectations with him, though I said the same about Waddle for Thursday and he had a very decent game, so I’m manifesting your success with this post.
 
Most underrated fant WR I keep thinking when will he regress and I bench him but proves me wrong week after week. Imagine he had a legit QB.
Regress? He was underperforming all year when measuring his xFP's. He's been fed tons of volume and finally broke out this week.

Crickets in here. Olave has been an awesome WR3.
 

Regress to the mean. xFP. If he starts to live up to his xFP, then he's "regressing" because "regression to the mean" means he scores an amount of points closer to his expected points, which means he's scoring higher than he had before comparatively because his xFP was so much greater compared to his actual points.

The thing here is the difference between actual points and expected points. You "regress" when, if you have underperformed, you score closer to the expected point total than you have previously.

Say Receiver A scores 8 actual points out 15 xFP for Week 1, 7 actual points out of 17 xFP for Week 2, and 14 actual points out of 15 xFP for Week 3. For FF scoring, we only count 8, 7, and 14 points, right? But Receiver A, if we look at the average receiver when given the situation and whatever factors you're holding constant, was expected to score 47 points. So our Receiver A underperformed an average receiver in history by a total of 18 points.

But even though he underperformed in every week, the last one, Week 3, is "regressing" because it's showing less of a difference between his actual and expected point total. -7, -10, -1

Important: "Regression to the mean" in fantasy football terms when discussing expected points is usually a term of art when considering cumulative expected points because it's not really a true mean you're regressing to in a mathematical sense, but a decrease in the variance of actual points to expected points from week to week. The baseline is moving and his xFP can change for a variety of reasons, so using a true mean of xFP isn't always a good idea. We assume it will be relatively constant, but there are instances where it isn't.

So, in sum, it means Olave scores more actual points than in previous weeks when compared to his xFP for that week. He is therefore "regressing" but scoring more because he is closer to what you'd expect given his xFP and given his prior underperformance in actual points vs. xFP.

I hope that clears it up. Goggins means Olave will, in the future, score more points than he had previously, which will be "regressing." But even though he is scoring more points, he will be on Goggins's bench because Goggins can't trust him (not due to underperformance in relation to xFP, but due to performance compared to alternatives scoring more actual points in Goggins's WR spot) to score more actual points than Goggins's other WRs score actual points.

eta* I keep editing because of the way the industry uses "regress" and how FF players are supposed to see it.
 
Last edited:

Regress to the mean. xFP. If you start to live up to your xFP you're "regressing" because "regression to the mean" means you start to creep towards the average, which means he's scoring higher than he had because his xFP is so high relative to his actual points.

It means he's scoring closer to what you'd expect given his xFP
Great explanation.

So...what's an xFP?

yeah, that's an awesome question. It's xFP, or expected fantasy points. It's a number produced by fantasy analysts that purports to show that with certain similar variables, this is what we should expect a player to score given his situation (often times his targets, their location, etc.)

I hadn't ever looked too hard into the sausage making because I stuck with basics, but I have, over the past months, begun to do exactly that which I didn't do before. I went down a rabbit hole with baseball and cross-positional comparisons of WAR (Wins Above Replacement) and you can find that over in the FFA 2025 Baseball Thread. I'll link to it. I am NOT a quant or math guy, so my investigations are limited. But I can tell you when they're making a bad assumption about a sport and incorporating it into a model. Here is where I put WAR from Fangraphs on blast, but I am not perfect and am still developing my arguments. I know cross-positional WAR is just making way too many assumptions and it gets circular in its logic. It's weird and abstract, but I do not think the way they measure WAR across positions for defense is appropriate or correct. They also do not take into account the scarcity of replacement offensive production at each position in order to come up with a "replacement' level player. (I don't think.)

Do not take this as gospel nor finalized. It's just showing where I'm sort of headed these days.

 
Last edited:

Regress to the mean. xFP. If you start to live up to your xFP you're "regressing" because "regression to the mean" means you start to creep towards the average, which means he's scoring higher than he had because his xFP is so high relative to his actual points.

It means he's scoring closer to what you'd expect given his xFP
Great explanation.

So...what's an xFP?

yeah, that's an awesome question. It's xFP, or expected fantasy points. It's a number produced by fantasy analysts that purports to show that with certain similar variables, this is what we should expect a player to score given his situation (often times his targets, their location, etc.)

I hadn't ever looked too hard into the sausage making because I stuck with basics, but I have, over the past months, begun to do exactly that which I didn't do before. I went down a rabbit hole with baseball and cross-positional comparisons of WAR (Wins Above Replacement) and you can find that over in the FFA 2025 Baseball Thread. I'll link to it. I am NOT a quant or math guy, so my investigations are limited. But I can tell you when they're making a bad assumption about a sport and incorporating it into a model. Here is where I put WAR from Fangraphs on blast, but I am not perfect and am still developing my arguments. I know cross-positional WAR is just making way too many assumptions and it gets circular in its logic. It's weird and abstract, but I do not think the way they measure WAR across positions for defense is appropriate or correct. They also do not take into account the scarcity of replacement offensive production at each position in order to come up with a "replacement' level player. (I don't think.)

Do not take this as gospel nor finalized. It's just showing where I'm sort of headed these days.

Thanks, RockGPT!

Personally I think all "over expectation" "statistics" relating to magic football are 🐴 🙂<---- that's what this MB outputs when you use the poop emoji.
 
Thanks, RockGPT!

You're welcome! I didn't use AI for that post at all, if that's the implication. I just know it.

I did use it before to make sure that when somebody uses "regressing" in casual conversation regarding FF, which should be short for "regressing to the mean," that they're using it to invoke "regressing to the mean" as a sort of colloquial rubric instead claiming that there's an actual mathematical average we're comparing the number in question to. If they were being truly literal about it, they wouldn't write about it the way they do, but I wanted to double check and source it.

I actually ran Olave's numbers, but I think AI was wrong so I ignored it. It's often just wrong and you have to look at its sources and see if it's reading those correctly.

For example, I had a discussion with it recently about how AI was reading something and interpreting it. It was clearly wrong. You cannot just use this stuff blindly. It was about baseball WAR and it was taking a heading of a paper and using it incorrectly regarding what it meant for the paper. It was just wrong. So if you use it blindly, you'll be wrong and you'll steer people wrong, and I don't know how they're using it in businesses and in law. It's irresponsible. It's still just a search engine/aggregator with probability baked into language. I have the exchange if you're interested. You can PM me. It shows how it can botch a reading and then interpret a mathematical formula completely incorrectly. And it will regurgitate it back to you sweetly and innocuously.

Personally I think all "over expectation" "statistics" relating to magic football are 🐴 🙂<---- that's what this MB outputs when you use the poop emoji.

Fair. You really have to understand what the model is telling you. Dude, if you're interested, check the thread I just started about Rushing Yards Over Expectation. That didn't come from AI, either. It came from Sumer Sports and other sources. I read it because there is so much out there that quant guys are doing, but I remember baseball and their SABR revolution, and how in the '90s they were making the WILDEST assumptions about the game and human nature. They're actually arguing about momentum right here on these boards over in the "coaching disagreement" thread, and the quant guys aren't right and the momentum advocates aren't right either, IMO. Actually, I shouldn't say they're right or wrong. I don't want to involve myself but I think they're arguing from two different perspectives and it's interesting to see.

But there is no doubt that people are using "rushing yards over expectation" by the NFL Next Gen Stats folks incorrectly all over Twitter, and some darn good analysts are doing it. It is NOT a comparative stat. The inputs into the model are different than what people think.

I think that inquisitiveness is called for in this environment and you are right to be skeptical about certain stats. I think CPOE (completion percentage over expectation) is okay but I need to really look into it (how are they accounting for pass rushing and clean pockets, etc.), but you may have your reasons for not digging it.

Football is totally unlike baseball in that there's no isolated event of two players like a pitcher and a batter. Nothing interferes with the pitcher/batter dynamic enough to prevent a person from drawing tons of inferences and measurements with cold-*** stats as their guide. But football is different. There are twenty-two interlocking men in football and it prevents you from isolating anything or saying much about causation of anything. That said, no good quant guy ever says anything but that football is much more difficult with a unique set of challenges.

And that's about it. I've always looked at the advanced stats but never the ones that seemed to be problematic from the jump. It's good to know the model and its assumptions before either embracing it or dismissing it, frankly.
 
dreams of olave and josh allen.

It does feel like Buffalo could be trending towards getting Allen another legit WR but that could change if they get back on track but agree Olave would be a great fit there...unless they are concerned about his durability I'm not sure if dealing Olave would be a smart move for the Saints...big picture I get it because they are in a rebuild but whether it is Rattler, Shough or a rookie next year they will be developing a young QB and losing Olave will only make that process more difficult...if he did become available I would love for the Pats to get aggressive in going after him...would be a perfect fit with Maye.
 
What is his dynasty value right now? Worst team in my league has him. I could use a little extra juice for a playoff push.
 

Users who are viewing this thread

Back
Top