And yet he wasn’t even the unluckiest player in the league this year. The weird thing about my season is that I didn’t actually have a close game all year until now. I either blew out my opponent or got blown out. I think my closest game was a 12 point margin.
It was a weird year all around. I was trying an on the fly rebuild, and certainly never thought I had the “to beat” team. But I had the depth and the timing worked out for me, and here we are.
The arrival of Jaxon Smith-Njigba and Trevor Lawrence, the resurgence of Johnny Taylor, the roller coaster ride of Jahmyr Gibbs, the cratering of Justin Jefferson and Brian Thomas Jr., and the steadfastness of Josh Allen were all interesting storylines. And all of them show just how much fantasy football is a week to week game.
Drake Maye scored 5 TDs this week so I’ve got the top 2 scoring QBs.
I have one RB, and no WRs, but at least I have kicker locked up and the 2nd best QB riding my bench. Optimal team construction.
I was thinking Loveland was a bit disappointing though the best TEs often don’t break out until year 2 but he racked up a bunch of solid games and shockingly was the Bears most prolific receiver with 713 yards, which is kind of shocking since they’re stacked for TE targets. So that seems pretty good.
Also, Kyle Pitts is a free agent. He’s going to sign with Kansas City and become the new Travis Kelce for the next 10 years and I will have to start considering starting 2 TEs like old times since I have no receivers.
I (unfortunately) now have to pay attention to the Cardinals coaching search and QB situation as Michael Wilson has finally turned into an asset
If the posted standings are correct I have the 2nd (mine) the 4th (Justin’s) and 9th picks (Jules).
Probably not the greatest draft class to load up on, but I will hopefully have a chance to start stacking up some players for a change.
I was really happy about my Jeanty-Judkins swap right up until the point where Judkins’ knee exploded.
Hey guys!
You know how Jules makes really thorough analysis and spreadsheets and all that?
And Ellis actually wrote a program years ago to parse yahoo data to give us interesting data about managerial decisions and max potential points?
Well I did a small, lazy fraction of that work! I vibe coded an app that scraped our league history into a format I can interrogate with random questions.
For instance, want to know what players were owned by the same team for the most consecutive years?
| Rank | Player | Manager | Years | Span |
|---|---|---|---|---|
| 1 | Jets D/ST | Ellis Dee | 17 | 2009–2025 |
| 2 | Packers D/ST | Retrovertigo | 14 | 2009–2022 |
| 3 | Tom Brady | Retrovertigo | 14 | 2009–2022 |
| 4 | Keenan Allen | Hamlet | 13 | 2013–2025 |
| 5 | Chiefs D/ST | Omniscient | 13 | 2013–2025 |
| 6 | Aaron Rodgers | SenorBeef | 12 | 2009–2020 |
| 7 | Philip Rivers | Hamlet | 12 | 2009–2020 |
| 8 | Justin Tucker | Hamlet | 12 | 2013–2024 |
| 9 | A.J. Green | Retrovertigo | 12 | 2011–2022 |
| 10 | Frank Gore | Justin Bailey | 12 | 2009–2020 |
If we count D/ST, then Ellis has owned the Jets D/ST for the entire league history. If we’re talking about individual players, it’s Tom Brady held by Retrovertigo for 14 years.
It’s like our own little stats analysis program. Anyone have fun questions I can run through it?
I asked it: give me the top 10 pairings that played the most games against each other along with their record in those matchups. This is determined by longevity in the league but also because division rivals play twice per year.
| Rank | Matchup | Games | Record |
|---|---|---|---|
| 1 | Hamlet vs Retrovertigo | 34 | Hamlet leads 25-9 |
| 2 | Dale vs RNATB | 34 | RNATB leads 22-12 |
| 3 | Ellis Dee vs Justin Bailey | 34 | Tied 17-17 |
| 4 | Omniscient vs Retrovertigo | 32 | Retrovertigo leads 17-15 |
| 5 | Hamlet vs Omniscient | 32 | Hamlet leads 24-8 |
| 6 | RNATB vs SenorBeef | 28 | SenorBeef leads 15-13 |
| 7 | Dale vs SenorBeef | 26 | SenorBeef leads 15-11 |
| 8 | Petey vs RNATB | 24 | RNATB leads 13-11 |
| 9 | Dale vs Petey | 24 | Petey leads 15-9 |
| 10 | Petey vs SenorBeef | 22 | SenorBeef leads 12-10 |
All time win loss records, ranked by winning percentage:
| Rank | Manager | Record | Win % | Seasons |
|---|---|---|---|---|
| 1 | Stringer | 83-43 | .659 | 9 |
| 2 | Jules Andre | 91-49 | .650 | 10 |
| 3 | SenorBeef | 142-96 | .597 | 17 |
| 4 | Peteys Part 2 | 40-30 | .571 | 5 |
| 5 | VarlosZ | 69-57 | .548 | 9 |
| 6 | Petey | 90-78 | .536 | 12 |
| 7 | Hamlet | 127-111 | .534 | 17 |
| 8 | RNATB | 125-113 | .525 | 17 |
| 9 | Justin Bailey | 120-118 | .504 | 17 |
| 10 | Furt | 46-52 | .469 | 7 |
| 11 | Ol’Gaffer | 52-60 | .464 | 8 |
| 12 | Ellis Dee | 109-129 | .458 | 17 |
| 13 | Spiritus Mundi | 25-31 | .446 | 4 |
| 14 | Retrovertigo | 99-139 | .416 | 17 |
| 15 | Dale | 96-142 | .403 | 17 |
| 16 | Omniscient | 88-136 | .393 | 16 |
| 17 | Overly Sentimental | 21-35 | .375 | 4 |
| 18 | Mad Hermit | 5-9 | .357 | 1 |
I had a hard decision to make. Am I going to filter out stats that make Jules look good? And I decided no, let’s just see the fun stats.
Biggest blowouts in league history:
| Rank | Season | Week | Winner | Score | Loser | Score | Margin |
|---|---|---|---|---|---|---|---|
| 1 | 2023 | Wk 4 | Jules Andre | 199.75 | Retrovertigo | 71.06 | 128.69 |
| 2 | 2020 | Wk 4 | Retrovertigo | 203.19 | Dale | 81.50 | 121.69 |
| 3 | 2018 | Wk 9 | Justin Bailey | 160.15 | Petey | 52.53 | 107.62 |
| 4 | 2011 | Wk 13 | Stringer | 185.45 | Omniscient | 78.50 | 106.95 |
| 5 | 2022 | Wk 6 | Jules Andre | 152.22 | Justin Bailey | 47.72 | 104.50 |
| 6 | 2023 | Wk 10 | RNATB | 158.57 | Omniscient | 54.72 | 103.85 |
| 7 | 2023 | Wk 1 | Jules Andre | 164.60 | Ellis Dee | 63.80 | 100.80 |
| 8 | 2018 | Wk 10 | Hamlet | 153.45 | Petey | 54.10 | 99.35 |
| 9 | 2021 | Wk 7 | Jules Andre | 138.10 | Omniscient | 43.88 | 94.22 |
| 10 | 2018 | Wk 14 | Jules Andre | 167.60 | Overly Sentimental | 74.10 | 93.50 |
Top 10 highest scoring seasons:
Here are the 10 highest-scoring regular seasons in league history:
| Rank | Manager | Season | Points For | Record | Final Standing |
|---|---|---|---|---|---|
| 1 | Jules Andre | 2022 | 2,003.89 | 12-2 | #4 |
| 2 | Jules Andre | 2023 | 1,966.95 | 11-3 | #2 |
| 3 | Hamlet | 2018 | 1,959.15 | 10-4 | #3 |
| 4 | Hamlet | 2009 | 1,935.30 | 8-6 | #3 |
| 5 | Stringer | 2011 | 1,921.40 | 11-3 | #2 |
| 6 | Hamlet | 2019 | 1,901.80 | 9-5 | #1 |
| 7 | RNATB | 2018 | 1,890.19 | 11-3 | #4 |
| 8 | VarlosZ | 2015 | 1,885.04 | 9-5 | #1 |
| 9 | Justin Bailey | 2013 | 1,876.51 | 11-3 | #4 |
| 10 | Stringer | 2012 | 1,859.98 | 11-3 | #4 |
Here’s every manager ranked by playoff consistency:
| Rank | Manager | Playoffs | Seasons | Rate | Titles | Runner-up |
|---|---|---|---|---|---|---|
| 1 | Jules Andre | 8 | 10 | 80.0% | 1 | 2 |
| 2 | Stringer | 7 | 9 | 77.8% | 2 | 2 |
| 3 | SenorBeef | 12 | 17 | 70.6% | 3 | 3 |
| 4 | Hamlet | 10 | 17 | 58.8% | 4 | 2 |
| 5 | VarlosZ | 4 | 9 | 44.4% | 3 | 1 |
| 6 | Ellis Dee | 6 | 17 | 35.3% | 1 | 1 |
| 7 | Petey | 4 | 12 | 33.3% | 0 | 1 |
| 8 | Justin Bailey | 5 | 17 | 29.4% | 0 | 0 |
| 9 | RNATB | 5 | 17 | 29.4% | 1 | 2 |
| 10 | Furt | 2 | 7 | 28.6% | 0 | 2 |
| 11 | Peteys Part 2 | 1 | 5 | 20.0% | 1 | 0 |
| 12 | Retrovertigo | 2 | 17 | 11.8% | 1 | 0 |
| 13 | Omniscient | 1 | 16 | 6.2% | 0 | 0 |
| 14 | Dale | 1 | 17 | 5.9% | 0 | 1 |
| 15 | Ol’Gaffer | 0 | 8 | 0.0% | 0 | 0 |
| 16 | Overly Sentimental | 0 | 4 | 0.0% | 0 | 0 |
| 17 | Spiritus Mundi | 0 | 4 | 0.0% | 0 | 0 |
| 18 | Mad Hermit | 0 | 1 | 0.0% | 0 | 0 |
And.. I just want to say… it’s really remarkable that we’ve had a league that has ran 17 years and who knows how many more. Dynasty was only starting to become trendy when we started this league, which means we were among the first leagues to seriously try it. And I bet 99%+ of those leagues created in 2009 have dried up by now. We’re the elite endurance athletes of the dynasty fantasy football world, and I want to thank all of you that play with us for sticking in there year after year even through rough times and making this such a durable and reliable league. We’ve only had 6 members turn over in 17 years. That’s amazing for a fantasy football league. And it’s not a money league. It’s not a league of real life friends. We’re just a bunch of random dudes on a message board who have bonded over Ellis Dee owning the Jets Defense for 17 years.
Those stats are awesome.
I had been feeling pretty good about my team for a couple years now, but I see I still have not caught up to furt in terms of winning percentage. Ouch.
So let’s see, 6 total playoff appearances for me in 17 years…
2025 made the playoffs (3rd)
2024 made the playoffs (4th)
2023 made the playoffs (3rd)
…means I was a dreadful 3 playoff appearances in my first 14 seasons. Okay, yeah, I guess I can see why I still haven’t caught furt yet.
How about that parity between me and Justin, eh? 17-17. I like it.
What AI did you use? How good is the data?
Can you get game-level stats, say asking which #1 overall draft pick has stared the must games for the team that drafted him? Can you analyze historic trades to determine which side won?
So I used Claude. Claude Code built a python script that used yahoo’s API to pull data from our league - matchups, managers, drafted players, scores - all stuff that you could get yourself by looking through yahoo’s league history function but pulling it all efficiently at once. Then it put it all in a spreadsheet (I can send you guys the spreadsheet if anyone is interested). Then I can use Claude chat to query the spreadsheet - he understands the format.
So I can easily ask it any questions that you could hypothetically answer yourself if you just looked through all of yahoo’s data. If you wanted to know “what was the biggest blowout in league history” you could go through every year in league history, look at the scores, figure it out yourself - it would just be a hassle. This obviously makes that process very easy.
You could ask it questions that rely on career trajectory and stuff like that, but that isn’t in the spreadsheet. He’d have to do a web search or use his internal knowledge. That would be a bigger computing ask than just querying the spreadsheet. So you’d probably want to ask it more specific, narrow questions for that kind of thing - “rank everyone’s draft picks by how well they did in their careers in the NFL” is conceptually possible but it would take a lot of work for the LLM (and probably use up half my subscription usage for that week).
But any question that is basically already in the spreadsheet is trivial. He was able to tell me who were the longest kept players just be looking to see what players were in the same team’s “draft” every year. (I explained to him our keeper system, that there were a few rookies but mostly the “draft” was veteran players we’re keeping) - and he went in and flagged every player’s first year in the league to be able to figure out who the actual rookies were versus the veterans. That’s how I was able to generate this data, for example:
Here’s the positional breakdown of each manager’s first 3 rookie picks every year (2010–2025):
| Manager | Picks | RB | WR | QB | TE | K |
|---|---|---|---|---|---|---|
| Furt | 18 | 61% | 28% | 11% | 0% | 0% |
| Justin Bailey | 48 | 50% | 29% | 6% | 8% | 6% |
| Overly Sentimental | 10 | 50% | 0% | 20% | 20% | 10% |
| Jules Andre | 30 | 43% | 47% | 3% | 7% | 0% |
| SenorBeef | 48 | 42% | 44% | 6% | 6% | 0% |
| Spiritus Mundi | 12 | 42% | 58% | 0% | 0% | 0% |
| Dale | 47 | 40% | 23% | 15% | 17% | 4% |
| Omniscient | 48 | 40% | 35% | 19% | 4% | 0% |
| RNATB | 48 | 38% | 33% | 15% | 12% | 0% |
| VarlosZ | 24 | 38% | 38% | 17% | 8% | 0% |
| Retrovertigo | 48 | 35% | 44% | 8% | 12% | 0% |
| Petey | 32 | 34% | 41% | 16% | 6% | 3% |
| Ellis Dee | 46 | 33% | 37% | 15% | 15% | 0% |
| Ol’Gaffer | 24 | 33% | 46% | 8% | 12% | 0% |
| Stringer | 24 | 33% | 46% | 0% | 21% | 0% |
| Hamlet | 48 | 31% | 33% | 15% | 19% | 2% |
come to think of it, I don’t have trading data in the spreadsheet. I wouldn’t be surprised if I could - if refine the api pull app to look at league actions like pickups, drops, and trades - assuming yahoo stores them. Right now it’s just - here are all the games, the scores, the players we drafted, which manager (human owner) was which team name, which years they played.
I could also probably get individual player scores in each matchup from the yahoo API - it would just require tweaking what the python app is seeking. Currently, I don’t have the data to say “rank the 10 highest scoring games by QBs in league history” but I don’t think there’s any reason I couldn’t as long as yahoo makes that information accessible through the API, which it probably does. I might indeed refine the app to pull that data sometime in the future.
I should add that I don’t think me being commissioner is a necessary step in all of this. I think just being a member in a league allows you to pull data from that league through yahoo’s API, so you could do something similar yourself if you wanted. Or I can just send the spreadsheet and you can interrogate that however you like.
Actually, now that I think about it, that extra data I didn’t yet put into the spreadsheet would be really interesting. If we saw the scores of every player every week, and what position they started in (RB, RB/WR, Flex, Bench), we could ask questions like “what were the highest scoring players in league history that were left on the bench by their manager” or “what are the top 10 scoring WRs overall in league history over their entire careers” which are interesting questions.
I might add that data this week, or I might work on that next week - I’ve been burning through my usage budget pretty quickly this week.
Thanks for the info, @SenorBeef. It’s fascinating. To have this league survive 17 fucking years is amazing, especially considering it’s based on an obscure message board. This is my second favorite league, and it remains the #1 reason I pay any attention to college football or the draft anymore.
This info defintely makes me want to keep Keenan Allen, even if he is entirely irrelevant from fantasy perspective, just so he beats Tom Brady.
Any chance you could easily find out the positional percentages of 1st round draft picks? My perception is that this league is highly skewed to picking RBs early, more so than any league I’ve ever been in. I’m wondering if the stats bear that out.
So answering your question was easy:
Here’s the first round positional breakdown across all rookie drafts (2010–2025):
Overall: 192 first round picks
- RB: 87 (45.3%)
- WR: 77 (40.1%)
- QB: 19 (9.9%)
- TE: 7 (3.6%)
- K: 1 (0.5%)
- DEF: 1 (0.5%)
But then I saw… a kicker and a defense in the first round? that can’t be right. there must be some sort of record keeping error.
So I did some investigation. The spreadsheet confirms those picks were real - or at least they existed alongside other plausible draft picks. So I dug out the old SDMB FF Dynasty League: Year Two to investigate. I forgot this - it was 16 years ago - but we didn’t have the in-thread draft system in year 2. We entered all our keepers into yahoo and then had a live draft through the app. And some players couldn’t attend the live draft, so they pre-ranked their list, but yahoo will always auto draft empty starting spots before drafting players from your pre-rankings, so these two owners actually drafted a kicker and defense in the first round. It was probably one of the reasons we moved to the (much superior) thread draft.
I’ve been vague on which owners drafted which kickers/defense. I don’t want to steer up painful 16 year old memories haha. I think we debated coordinating another draft to rectify this problem but there were logistical and other problems with it. Maybe we threw them some FAAB dollars or future draft picks, I can’t remember.
Funny to see Varlos (I miss Varlos) happy that we retained all our managers into year 2. I hope he knows, whatever planet he’s on, that we’re still alive and kicking in year 17.
Yes, please. I’ll PM you my email.
I asked Claude to alter the numbers so that your team looks a little worse in every regard and then sent you that version of the spreadsheet.
(I didn’t actually do that)
(probably)
Did you ask Claude to ingest all the league year threads on SDMB? Might uncover some fun insights.
That’s an interesting idea. Haven’t tried that yet. I wonder how much usage it would cost to do something like that. Maybe I could strip out a lot of the redundant text first - like edit out the parts where people quote the draft results over and over again as we go along in the drafts.
What sort of ideas would you want to interrogate the old league text for?
See when people made predictions in the thread and how they turned out.
Capture rule changes and bylaw changes that were agreed to but never officially recorded as a bylaw in some master document that may have been forgotten over 17 years.
Comments about specific draft picks and how it turned out.
See if it can figure out the reason for Varlos’ mysterious disappearance. Maybe he left coded messages about being abducted by aliens.
Maybe capture what trades were made and what people were saying about them at the time - they’d probably be interested in retrospect. We were probably worried about a bunch of unbalanced-looking trades that broke our predictions entirely.
Calculate Hamlet’s asshole factor and how it varies across the years and even within the season. Is it modulated by his win-loss record in every year?