Projecting future 1st placement with good probability

[Jrblaha]

I’m seeing a lot of posts where people are saying early, mid, late 2025 or even two years out (2026).

For the sake of this post we’ll define predicting good as getting the right tier (early, mid, late) over 50% of the time. We can adjust that later if needed, but seems like we need a starting point if we’re gonna discus doing this “good.”

Are there people out there guessing correctly the majority of the time with decent sample sizes? (year after year after year etc).

My longest league has been around maybe 10 years and I’ve predicted the correct champ like once. Sometimes they don’t even make the playoffs. There’s so many factors that go into how a team does, injuries, trades, facing high/low scoring teams, etc. The team that won it all for us this year, was predicted by myself and the team owner to be a non-playoff team.

My league is a classic 12 team league, division winners and then two best make the playoffs, byes for best two divisional winners, one game elimination playoff games. I get different setups possibly changing likelihood of predicting standings.

Outside of noticing an older team, with injury concerns, and lack of depth and draft capital to make changes, my predictions is off bad enough I’m not comfortable projecting beyond simply playoffs or not-playoffs, and even then, I’m not confident.

Anyone out there with factors that seem to give them good probability predicting final standings (and draft placement)?

[81-]

Nobody knows. This year I traded for a first thinking it would be top 3. It wound up being 11.

[tstafford]

Nobody knows. This year I traded for a first thinking it would be top 3. It wound up being 11.

Yep. I swapped 2nd round picks with a team thinking they'd be far worse. I ended up giving the 2.04 for what became the 2.10.

[Shcritters]

I’m seeing a lot of posts where people are saying early, mid, late 2025 or even two years out (2026).

For the sake of this post we’ll define predicting good as getting the right tier (early, mid, late) over 50% of the time. We can adjust that later if needed, but seems like we need a starting point if we’re gonna discus doing this “good.”

Are there people out there guessing correctly the majority of the time with decent sample sizes? (year after year after year etc).

My longest league has been around maybe 10 years and I’ve predicted the correct champ like once. Sometimes they don’t even make the playoffs. There’s so many factors that go into how a team does, injuries, trades, facing high/low scoring teams, etc. The team that won it all for us this year, was predicted by myself and the team owner to be a non-playoff team.

My league is a classic 12 team league, division winners and then two best make the playoffs, byes for best two divisional winners, one game elimination playoff games. I get different setups possibly changing likelihood of predicting standings.

Outside of noticing an older team, with injury concerns, and lack of depth and draft capital to make changes, my predictions is off bad enough I’m not comfortable projecting beyond simply playoffs or not-playoffs, and even then, I’m not confident.

Anyone out there with factors that seem to give them good probability predicting final standings (and draft placement)?

Every year in the preseason I create a spreadsheet and pull up predicted points for each anticipated starter on each roster. I then add up the points and rank the teams. With this strategy I’ve been able to predict pretty well… have regularly gotten a top 3 pick from one team the last 3 years.

KTC team ranker isn’t very reliable… they are more about total dynasty value rather than a one-year window.

Yes, it can be unpredictable, but I’d say I am able to guess early/mid/late a year in advance >50% of the time. But in that league it is pretty clear which teams are competing, which are rebuilding, and which are in the middle.

2 years out… different story. Good luck with that unless you have absolute killer or bottomed out teams.

[TheTroll]

I went from worst to 2nd in a single year. Honestly we’d like to think we know but any draft pick that is not the current draft year is a gamble. It is why they all are discounted (and should be). Future picks have value but just like a scratch off lottery ticket.

Where I think we get sidetracked is focused on the team at the time of the trade and not realizing. That so many other future moving parts will play a decision that is out of our control. Unless you are trading with a team that is in a deliberate rebuild, we have no idea. Catch 22 is that almost always they own their own picks.

The key is build up draft capital with the only expectation that they can be used to consolidate when needed. You can get lucky too but that is just what it is…luck.

[SoftwoodGrampian]

I have found it to be a complete crapshoot

[killer_of_giants]

buy supposedly late 1sts, at late 1st prices. that way, you'll only be positively surprised.

[TheTroll]

buy supposedly late 1sts, at late 1st prices. that way, you'll only be positively surprised.

I love that!

[BabyChark23]

buy supposedly late 1sts, at late 1st prices. that way, you'll only be positively surprised.

Good advice! I’ve done this the last two years with mixed results. Last year it resulted in the 1.02. This year it resulted in the 1.08. But I paid late 1st prices for both picks, so they were both Ws. If you target teams that you know are delusional or that have weak depth despite strong starters, it can turn out a win for your team more often than not.

[TheTroll]

buy supposedly late 1sts, at late 1st prices. that way, you'll only be positively surprised.

Good advice! I’ve done this the last two years with mixed results. Last year it resulted in the 1.02. This year it resulted in the 1.08. But I paid late 1st prices for both picks, so they were both Ws. If you target teams that you know are delusional or that have weak depth despite strong starters, it can turn out a win for your team more often than not.

…and if you don’t know what owner is the delusional one, then it’s you!

[jordanzs]

I’m pretty good at short selling certain owners based upon their tendencies (or lack of tendencies). Sometimes they just don’t spend enough time on the waiver wire during the season so they miss out on pieces to keep them afloat for playoff contention. Some owners always trade away vets for rookie picks. I love when one of those owners gets excited and snap accepts my offer of “my” first and second for “their” first.

[MacDaddy123]

I pretty much just laugh at projections more than 1 year out.
I see it done a lot by people trying to argue how lopsided a trade is.
They will project one side to be getting early 25 and early 26 1sts, while projecting the other side to be getting late round picks.

It's comical to me. I mean you can make a trade calculator say anything you want it to say by loading it with false information.
I value all 25 and 26 picks today as mids. I can't start gauging early or late until about 6/7 months before the NFL Draft.

[brward]

Yup I've gone from the most points in one year to the fewest the next year just based on guys have good vs bad years and injuries. No one really wanted my projected late first until it ended up as the 1.01/saquon.

[Dynasty DeLorean]

Every year in the preseason I create a spreadsheet and pull up predicted points for each anticipated starter on each roster. I then add up the points and rank the teams. With this strategy I’ve been able to predict pretty well… have regularly gotten a top 3 pick from one team the last 3 years.

KTC team ranker isn’t very reliable… they are more about total dynasty value rather than a one-year window.

Yes, it can be unpredictable, but I’d say I am able to guess early/mid/late a year in advance >50% of the time. But in that league it is pretty clear which teams are competing, which are rebuilding, and which are in the middle.

2 years out… different story. Good luck with that unless you have absolute killer or bottomed out teams.

This sounds like a cool idea.

In my experience, it's very hard to project just by eye-balling. The stronger/older the team, the further out I try to buy future picks if possible. So if you're buying a 1st from a contender, if given the choice, I'd rather have a 2026 1st instead of a 2025 1st. Just gives it that 1 extra year for a chance to fall apart.

I've had mixed luck with trading for future picks. I kept trading for a guys future 1st in my league. His roster had zero depth, and I mean zero. He could barely field a starting lineup. But all his starters had career years like 2 seasons in a row when I had his picks, and they never got injured. He would up making the playoffs both years. Meanwhile, half my starting lineup is out due to injuries all the time I feel like.

Sometimes I get lucky. I traded a producing RB at the time (Khalil Herbert) to a team who thought they were a contender. The pick ended up being the 1.02, and then I traded up to get the 1.01 (Bijan). Yeah.... that person never wants to trade with me anymore lol. I also traded 3 3rds and a 4th for a future 2nd this year, the 2nd being from a team that just went 9-5. Well, the guy earned the 1.01 and that 2nd ended up being the 2.01, which I just traded for Tank Dell. So trading for future picks can be lucrative. You just can't overpay assuming it will be early. As someone said before, value them all as late-1sts (unless the other team is truly horrific) and you'll never be disappointed.

[Shcritters]

Every year in the preseason I create a spreadsheet and pull up predicted points for each anticipated starter on each roster. I then add up the points and rank the teams. With this strategy I’ve been able to predict pretty well… have regularly gotten a top 3 pick from one team the last 3 years.

KTC team ranker isn’t very reliable… they are more about total dynasty value rather than a one-year window.

Yes, it can be unpredictable, but I’d say I am able to guess early/mid/late a year in advance >50% of the time. But in that league it is pretty clear which teams are competing, which are rebuilding, and which are in the middle.

2 years out… different story. Good luck with that unless you have absolute killer or bottomed out teams.

This sounds like a cool idea.

In my experience, it's very hard to project just by eye-balling. The stronger/older the team, the further out I try to buy future picks if possible. So if you're buying a 1st from a contender, if given the choice, I'd rather have a 2026 1st instead of a 2025 1st. Just gives it that 1 extra year for a chance to fall apart.

I've had mixed luck with trading for future picks. I kept trading for a guys future 1st in my league. His roster had zero depth, and I mean zero. He could barely field a starting lineup. But all his starters had career years like 2 seasons in a row when I had his picks, and they never got injured. He would up making the playoffs both years. Meanwhile, half my starting lineup is out due to injuries all the time I feel like.

Sometimes I get lucky. I traded a producing RB at the time (Khalil Herbert) to a team who thought they were a contender. The pick ended up being the 1.02, and then I traded up to get the 1.01 (Bijan). Yeah.... that person never wants to trade with me anymore lol. I also traded 3 3rds and a 4th for a future 2nd this year, the 2nd being from a team that just went 9-5. Well, the guy earned the 1.01 and that 2nd ended up being the 2.01, which I just traded for Tank Dell. So trading for future picks can be lucrative. You just can't overpay assuming it will be early. As someone said before, value them all as late-1sts (unless the other team is truly horrific) and you'll never be disappointed.

Following up on this... went and did the analysis to show the standard deviation for my pre-season compared to final rankings for this league for the last two years and the stddev was 2 and 2.48. For simplicity, let's use 2 as the number. Someone better at math than me is likely about to destroy my explanation... but this essentially means that if I project a team pre-season to be 3rd worst (1.03 pick) there's a 68% chance (1 stddev in either direction) that the pick will be between the 1.01-1.05. If the team was projected to be 5th best (8th worst, 1.08 pick) that means a 68% chance to be between 1.06-1.10. It isn't an exact science, but it is directional enough to be valuable in identifying 'top third middle third, and bottom third' likely teams. And once I identify those teams I layer in the owners' tendencies and rosters to try to identify potential trade targets. And over all of this is that many of those lower-ranked teams recognize they don't stand a chance of making the playoffs or winning a ship and are therefore unlikely to trade their future 1st.

Done differently... I took the difference between pre-season ranking for the top half teams and compared them against their finishing positions and they averaged at 3.75. Simply mathematically you'd expect it to be 3.5 (averaging 1+2+3+4+5+6 divided by 6). When doing the same for the bottom half the number was 9.25 (expected to be 9.5). I interpret this to say that 'on average, a pre-season projected top-half team will end the season in the top-half of the rankings and, on average, a projected bottom-half team will end the season in the bottom half of the rankings'. Even if you're looking to diving 'early vs. late' this is pretty useful.

Everyone always remembers the outliers, but the majority of the time it is useful for directional usage.