K.I.S.S. Model

Introduction:

Every year when it comes time to draft your fantasy football team, the internet floods with memes of the girlfriend, the cousin, or the totally clueless friend who picks based on name recognition alone. You know the type, they take the best quarterback on the best team because, well... that’s what makes sense to them.

For over 20 years, that meant drafting Tom Brady (and as a patriots fan, I get it). Now? It's Patrick Mahomes, Josh Allen, Lamar Jackson, or whatever other top QB that ESPN is hyping up that morning.

Now I am not telling you to go out and draft Tom Brady (Patrick Mahomes) at the 1.01, but is there some truth to the idea that drafting the best players on the best teams yields better results?

That is what this project will explore.

The League Format & The Data

One of the best things about fantasy football is how much data exists, years and years of stats, draft results, and rankings are all online because that’s where the game lives. So when it came time to pick a data source for this project, I had options: ESPN, NFL.com, Yahoo…

I went with none of them.

Instead, I chose Underdog Fantasy, a daily fantasy sports platform that runs some of the biggest best-ball tournaments every year, with prize pools reaching over $2 million.

Now, here’s the tradeoff, while sites like ESPN have over a decade of data, Underdog only goes back to 2022, so I’m working with three seasons (2022–2024). That’s not ideal for using past trends to predict future outcomes, but I still believe this is the best dataset for this project.

Why Underdog?

Because on Underdog, every draft costs money, anywhere from $3 to $500 per entry. Their biggest tournament, Best Ball Mania, costs $25 to enter and pays out $2,000,000 to the winner.

This matters. In free leagues, like those on Yahoo or ESPN, people might draft half-heartedly, auto-pick half the time, or take a defense in Round 5 in an attempt to fill out there lineup. On Underdog, there’s skin in the game, people take drafts seriously. And that means the Average Draft Position (ADP) data is sharper, more intentional, and likely more reflective of high-level strategic thinking.

You also get volume, tens of thousands of entries per year, many from some of the best fantasy players in the world. When you're paying real money to draft, you're not "winging it".

So while the Underdog dataset is smaller, I believe it’s more reliable and better aligned with the question I’m asking: should we be targeting good players on good teams?

How It Works

Underdog’s format is best ball, you draft a team once, and that’s it. No waivers, no trades, no lineup decisions. Each week, Underdog automatically sets your highest-scoring possible lineup. This rewards drafting well, which is exactly what I want to measure.

The scoring is Half PPR (points per reception), and starting lineups look like this:

1 QB
2 RB
3 WR
1 TE
1 FLEX (RB/WR/TE)

For this project, I used Underdog’s ADP data for draft position and matched it with end-of-season rankings from FantasyPros, using the same Half PPR format for consistency.

Data Collection

To begin, I pulled Underdog ADP data from FantasyPros, which keeps a historical archive by season. Luckily, this was downloadable in CSV format, so I was able to quickly upload each season (2022, 2023, and 2024) into Google Sheets and start cleaning the data to fit the needs of this project.

Sticking with FantasyPros, I also sourced end-of-season rankings from their site, specifically using Half PPR scoring to match Underdog’s format.

Finally, I pulled preseason win total lines from Pro Football Reference. If you're wondering why I used preseason over/under lines instead of a team’s actual win totals, that's a fair question. I did experiment with both, but here's the logic:

When you’re drafting a fantasy team, you don’t know how many games a team will win. You only have the expectation, which is what Vegas sportsbooks set. Including actual win totals would be hindsight bias, something drafters never had access to. And frankly, Vegas lines are usually more accurate than most of us anyway.

So ultimately, I stuck with preseason win expectations, because that’s what the average fantasy player is working with at the time of the draft.

Data Cleaning

Although the data was downloadable as a CSV file, it still required some cleaning before it could be used. The biggest issue was player names. When I first tried to match players’ ADP data with their end-of-season ranks, quite a few didn’t align.

A lot of the mismatches came from players with suffixes in their names, like “Jr.”, “Sr.”, or “III”. For example, Kenneth Walker III might be listed as Ken Walker in another dataset. These minor inconsistencies made a big difference when trying to merge data.

To fix this, I wrote some code that stripped out all suffixes and standardized names across sources. This brought my unmatched player count from 83 down to just 30. And of those remaining 30, all were irrelevant to this study, either injured, backups, or players who weren’t drafted in most leagues.

The next bit of cleaning involved team names. I had to make sure each team was labeled consistently across all data sources. The most common discrepancy came with the Jacksonville Jaguars, one source used "JAC," while another used "JAX." I standardized all team names to match a uniform 3-letter format to ensure a smooth merge later on.

Over-Performance by Team

The first bit of analysis I ran looked at player over-performance relative to ADP, specifically, how many players finished the season ranked higher than where they were drafted.

The results:

531 players over-performed their ADP
708 players underperformed

That’s a cool stat, but it doesn’t say much on its own.

So I decided to bring team context into the mix and see which teams had the most players who beat their ADP. Here were the top 10 teams ranked by average ADP overperformance:

CAR 5.909091
SEA 2.500000
IND -2.166667
DET -2.450000
WAS -4.542857
HOU -7.048780
PIT -9.323529
ATL -11.777778
MIA -13.578947
LV -13.692308

This is not what I expected to see. A team like Carolina leading the way?

But I have a couple of theories:

Bad teams have fewer fantasy relevant players, so the few that do get drafted are often later-round picks. That gives them more room to “beat” their ADP.
There's also a sort of “good team tax” baked into ADP already. The best players are getting drafted no matter what team they’re on, but when it comes to closer picks, people lean toward players on better teams. That inflates ADP for players on strong teams and suppresses it for those on weak teams, making it easier for those under-the-radar guys to overdeliver.

This heat map shows each team’s average overperformance, sorted by preseason Vegas win totals in descending order:

Once again, it reinforces the idea of a “good team tax.” Teams projected to be great are mostly shaded in blue, meaning their players underperformed ADP more often than not. Meanwhile, some of the lowest-ranked teams show up in red, indicating strong overperformance.

In other words: the better the team was expected to be, the more likely its players were drafted aggressively and failed to live up to it. Conversely, players on bad teams often slipped in drafts, giving them more room to beat expectations — especially when volume worked in their favor.

Do Good Teams Actually Help? (Vegas Win Totals)

As I mentioned earlier, I wanted to bring preseason win totals from Las Vegas sportsbooks into this study. These win projections are made before the season starts and represent how “good” we expect each team to be — which is what fantasy drafters are working with on draft day.

To simplify the analysis, I broke teams into three buckets:

Bad teams: 0–6.5 projected wins
Good teams: 7–9.5 projected wins
Great teams: 10+ projected wins

Just like in the earlier section where I looked at team-level overperformance, the results here might surprise you:

Once again, players from bad teams actually outperformed more than players from good or great teams. That seems counterintuitive, but it follows some of the same logic as before, and adds an important layer: talent vs. volume.

For example, imagine the WR2 on the Chiefs versus the WR1 on the Browns. Even if they’re similarly talented, the Chiefs receiver might face more competition for targets, while the Browns receiver could be the clear #1 option and dominate volume, even on a worse offense. In fantasy, volume often beats situation.

This next chart reinforces the same idea. Every dot represents a player, and the red line represents the line where a player’s final rank exactly matched their ADP.

Dots below the line = players who overperformed
Dots above the line = players who underperformed

If you look closely, the dots below the line skew darker, more greens, blues, and even purples, which means more players from lower-win teams are the ones exceeding expectations.

Meanwhile, players from top teams (shown in yellow and light green) cluster closer to or even above the line.

Added Stipulation: Controlling for Injuries

All these findings so far have been interesting, but something still felt off. My next thought was to filter the data down to only players who played 10 or more games. I felt that was a fair baseline. After all, injuries aren’t the same as underperforming, they’re unpredictable and often unavoidable. So I decided to remove most injury-shortened seasons from the study, and 10 games felt like a strong cutoff.

So, what changed?

Not much. In fact, the results became even stronger in favor of bad teams. Just like the earlier version of the heat map, the teams with the highest preseason win totals, the so-called “great” teams, are still mostly shaded in blue. Bad teams continue to show red, meaning their players consistently beat expectations.

This chart puts the differences into perspective. Bad teams now outperform good and great teams by an even larger margin than before. Controlling for injuries didn’t weaken the trend, it amplified it.

And finally, the filtered scatter plot tells the same story. Players from lower-win teams (shown in darker greens and blues) dominate the overperformance side of the red line (below it), while players from top teams cluster around or above the line, meaning they met or failed to meet expectations.

But here’s the thing: that conclusion still doesn’t feel totally logical. Should you really just draft players from bad teams to win your league? That can’t be the whole story.

It might work in some cases, but it clearly depends on position, depth chart, and volume, which is exactly what I’ll explore next, starting with the most important position in sports.

Quarterbacks

Earlier, I mentioned that drafting someone like Tom Brady at the 1.01 is a bad pick, and not just because he's retired. It’s because, although quarterbacks are the most important position in real-life football, that’s not the case in fantasy.

In most fantasy football leagues, you only start one quarterback. That means in a 12-team league, you only need 12 startable QBs, and there are usually more than that available. Some leagues use a Superflex format, where you can start two QBs, which increases their value dramatically. But this project is based on a standard 1-QB format, where quarterbacks are important, but not irreplaceable.

This first chart shows which rounds top-12 QBs (QB1s) were drafted. Surprisingly, the end of the draft produced the most QB1s. This lines up with what we’ve seen throughout the project: top QBs from high-win teams get drafted early and still perform well, but that Rounds 10–11 sweet spot is where some major value shows up.

Why? I see two explanations:

After the obvious top-tier QBs are drafted, the next group is a mixed bag. Some may be on good teams and others on bad ones, but ADP becomes more about perceived upside than certainty.
The most valuable QBs in fantasy often run the ball. Rushing boosts both the floor and ceiling of a QB. Think about Justin Fields on the Bears. Those teams weren’t good, and Fields wasn’t a great NFL passer, but he ran a ton, which made him a fantasy star. This rushing upside is often overlooked during the draft, especially when it comes from QBs on bad teams.

Of course, not all rushing QBs are on bad teams, think Josh Allen or Lamar Jackson, but you’re more likely to see lower-win teams lean on their QBs’ legs, simply because it's the most effective way for them to move the ball, and better teams may not want to run there QBs as much because they are focused on the playoffs and hopefully a Super Bowl, so keeping the QB healthy is the most important.

Do Good Teams Help at QB?

So I looked at the average projected win total for QB1s. The result, 9.03 projected wins.

That tracks. A good QB on a good team should have more scoring opportunities, better protection, and more weapons, all of which lead to more fantasy points.

Then I took it a step further: what happens when you split QB1s by whether their team had a projected win total of 9 or more?

QB1 Rate (Vegas ≥ 9 wins): 53.66%

QB1 Rate (Vegas < 9 wins): 28.57%

Players on strong Vegas teams are 1.88x more likely to finish as a QB1.

This confirms the logical thinking, good QBs on good teams perform well in fantasy. But there’s still nuance to consider.

Draft Strategy Implications

Just because high-win QBs have the highest hit rate doesn’t mean you have to draft them. Early QBs cost you a premium draft pick, which means passing on elite RBs or WRs. That’s where late-round QBs can shine.

This chart shows which rounds QB1s were drafted from, split by team strength. The blue bars represent QBs from teams with 9+ projected wins, and the red bars show QBs from teams with fewer.

From Rounds 2 to 5, every QB1 came from a high-win team, no surprises there. But in the late rounds, QB1s almost exclusively came from low-win teams. That’s huge. It means some drafters were able to land a QB1 in the final rounds while using their early picks on elite RBs or WRs, and those teams likely performed extremely well.

Final Thoughts

It all comes down to risk tolerance.

If you want safety, drafting an elite QB early is one of the most stable plays in fantasy, especially in your home leagues. But in large tournaments like Underdog Best Ball, the ceiling is everything. Many of the best players shoot for upside by taking dart throws on late-round QBs, often drafting 3 or 4 and only needing one to break out. If they hit, it’s a massive edge.

Running Backs

Running backs might be losing value in real-life NFL front offices, but in fantasy football, they’re still king. They touch the ball more often than most positions, rack up touchdowns on the ground and through the air, and most importantly, they offer consistent volume. Starting RBs don’t rotate nearly as much as WRs, so when you draft the right one, you usually know what you’re getting.

Another key difference from positions like QB? You start multiple running backs. So while it’s great to land a top-tier RB1, most rosters rely heavily on RB2s and even RB3s, which is why this section goes beyond just the top 12.

This chart shows what you'd expect: the vast majority of RB1s and RB2s are drafted before Round 5. No surprise here, these are typically your early-round targets, and managers chase running back volume aggressively at the top of drafts.

But interestingly, some RB1s and RB2s still emerge from the late rounds. Why? Because opportunity is king. These are often backups who benefited from an injury to the starter, leading to massive touch increases and breakout seasons.

Do Good Teams Help RBs?

Let’s bring Vegas into it, here are the average projected win totals:
RB1s: 9.027777777777779

RB2: 8.708333333333334

RB3s: 7.5

RB1s tied with QBs for the average win totals, and the 2nd highest total. That makes intuitive sense, better teams usually have better O-lines, more red zone opportunities, and positive game scripts that allow RBs to thrive.

Take Saquon Barkley for example. We watched him run into brick walls behind the Giants’ O-line for years, and while he was still productive, it was hard-earned. Fast forward to him joining Philadelphia, one of the best lines in the league, and he nearly breaks the single-season rushing record. Same player, new environment, massive jump.

Now here’s how win totals break down across tiers:

Of all RB1s:

61.11% came from teams with Vegas win totals ≥ 9

38.89% came from teams with Vegas win totals < 9

An RB1 is 1.57x more likely to come from a team with Vegas win totals ≥ 9 than from a team below 9.

Of all RB2s:

63.89% came from teams with Vegas win totals ≥ 8.5

36.11% came from teams with Vegas win totals < 8.5

An RB2 is 1.77x more likely to come from a team with Vegas win totals ≥ 8.5 than from a team below 8.5.

Of all RB3s:

61.76% came from teams with Vegas win totals ≥ 7.5

38.24% came from teams with Vegas win totals < 7.5

An RB3 is 1.62x more likely to come from a team with Vegas win totals ≥ 7.5 than from a team below 7.5.

So yes, RBs on better teams generally outperform those on bottom-tier squads.

Draft Strategy Implications

Running back is a position where you’re often rewarded for going early. Waiting on RB requires a lot of luck, hoping for injury upside, major depth chart shifts, or unsustainable touchdown rates. Those early RB1s like Christian McCaffrey or Saquon Barkley? They’re league-winners for a reason.

This chart makes a lot of sense in context. RB1s are relatively split between good and bad teams in the first few rounds , which reflects player talent > team situation. Think again of Saquon on the Giants, his talent kept him fantasy-relevant even when the team around him wasn’t.

But when he moves to the Eagles? That’s when everything breaks open. Better blocking, more red zone work, more scoring, it’s not that he’s a better player, he just has better support.

Here we see a more interesting pattern, especially in Rounds 15–17. Every RB2 taken in those late rounds was on a strong team.

Why? Because backups on good teams are worth betting on. If they get a shot, due to injury or role shift, they’ll be in high-efficiency situations. If you’re swinging for upside late in the draft, this could be a valuable strategy, target handcuffs on good offenses.

This one flips a bit. You see RB3s show up early (Rounds 1, 3, and 4) despite being on strong teams, which probably points to injuries. Even with the 10+ games filter, guys who barely hit that threshold might’ve missed key fantasy weeks, reducing their impact.

By the time you get to the end of the draft, most RB3s are coming from bad teams. That tracks, when you’re drafting dart throws on weak offenses, the ceiling is limited unless something unusual happens.

Wide Receivers

In Underdog drafts, many top drafters will tell you that the wide receiver position is the most important of them all. This is because you start more wide receivers than any other position. Securing elite WR production is crucial, they go early and often, and the position dries up fast.

Compared to the equivalent RB chart, you’ll notice a heavier concentration of WR1s going in Rounds 1–4. That’s no accident, elite fantasy wideouts are highly dependent on volume and quality quarterback play, and the ones in those situations go early.

Late-round breakout WRs happen far less frequently than with running backs. There are simply more WRs on the field, and roles are often more clearly defined. If Justin Jefferson gets hurt, the WR4 on the Vikings isn’t suddenly fantasy viable, Jordan Addison, who was drafted earlier, just gets more volume.

Do Good Teams Help WRs?

Bringing in the Vegas win totals, here’s how WRs break down by tier:

WR1s: 9.07

WR2s: 8.46

WR3s: 8.73

WR4s: 8.81

WR1s, on average, come from the strongest teams. That checks out, WRs are arguably the most dependent fantasy position. They need a good QB, who in turn needs a solid offensive line, and ideally other competent WRs to prevent bracket coverage.

Let’s look at how win total tiers correlate by fantasy tier:

Of all WR1s:

58.33% came from teams with Vegas win totals ≥ 9

41.67% came from teams with Vegas win totals < 9

WR1s are 1.40x more likely to come from strong Vegas teams.

Of all WR2s:

61.11% came from teams with Vegas win totals ≥ 8.5

38.89% came from teams with Vegas win totals < 8.5

WR2s are 1.57x more likely to come from strong Vegas teams.

Of all WR3s:

62.86% came from teams with Vegas win totals ≥ 8.5

37.14% came from teams with Vegas win totals < 8.5

WR3s are 1.69x more likely to come from strong Vegas teams.

Of all WR4s:

73.53% came from teams with Vegas win totals ≥ 8.5

26.47% came from teams with Vegas win totals < 8.5

WR4s are 2.78x more likely to come from strong Vegas teams.

Interestingly, WR1s have the smallest discrepancy. That somewhat contradicts the dependency argument but it reinforces the idea that volume is king. If a WR is talented and has a clear role, they can still finish high despite being in a bad offense.

Think about a situation like Malik Nabers last year. Everyone knew he was insanely talented and had the WR1 role locked up, even though the Giants were a mess. And yet, he got there. Why? 15+ target games. Pure volume.

The consistent trend across WR2s to WR4s, where the likelihood of success increases with team qualitysupports the notion that as talent decreases, dependence on offensive environment increases. That makes intuitive sense.

Draft Strategy Implications

Like RBs, the good WRs are going to go early and often, and like the data we just saw that could be contrary to how good the team is supposed to be.

Most players going in Rounds 1–3 are on high-win-total teams. After Round 4, there’s a noticeable cliff, at that point, you're drafting talented WRs in bad situations and hoping for target volume. There are only so many elite-situation WRs, and they get scooped quickly.

WR2s follow a similar pattern. High teams dominate the early and middle rounds. The drop-off begins around Round 7, after which you're mostly drafting WR2s from weaker teams. That may just be due to availability, the “safe” WR2s on good teams are already gone.

The WR3 chart is more chaotic. While players on good teams still dominate, the pattern is more dispersed. There’s less of a clean cliff like we saw with WR1s and WR2s, which makes sense. These are often ambiguous roles in real life and in drafts.

Nearly 75% of WR4s came from high-win-total teams, and the chart reflects that. It’s dominated by blue. So if you're looking to draft safe WR depth, you're better off targeting players on strong offenses.

Tight Ends

We come to the final position, another "onesie" spot like quarterback. For a long time, tight end has been one of the toughest positions to draft in fantasy football. Historically, there were only 4–5 reliable guys, and the rest were just dart throws.

This chart shows that a TE1 (top 12 tight end) can come from anywhere in the draft. That lines up with how the position works: there's usually a core group that’s expected to finish in the top 12 if they stay healthy, and after that, it’s wide open. You’re just hoping to hit on someone who pops.

Do Good Teams Help TEs?

Looking at average Vegas win totals, TE1s come from teams with an average of 8.79 wins. That’s pretty in line with what we saw at other positions, that 8.5 win mark seems to be a consistent threshold for fantasy ceiling.

Breakdown of TE1s by team win total:

Of all TE1s:

65.71% came from teams with Vegas win totals ≥ 8.5

34.29% came from teams with Vegas win totals < 8.5

A TE1 is 1.92x more likely to come from a team with Vegas win totals ≥ 8.5.

This continues a trend we’ve seen across all positions, players from better teams are more likely to finish as fantasy starters. TEs, like WRs, rely heavily on the rest of the offense, particularly quarterback play and red zone efficiency. The main difference is that WRs tend to be more explosive, which is why they hold more fantasy value overall.

Draft Strategy Implications

Tight end is one of those positions where you often have to choose, do you draft an elite TE early, or wait and take shots later?

Pairing a high-end QB and TE early isn’t usually optimal, you’re sacrificing valuable picks that could go toward stacking your RB and WR corps. And as we saw in the draft round chart, TE1s can emerge from anywhere, which opens up the door to find late-round value at the position.

This chart echoes what we saw in the win total breakdown. Early in the draft, TE1s almost exclusively come from strong teams. That’s when you’re drafting guys like Brock Bowers, Trey McBride, or George Kittle, players on decent offenses. Then, from the mid to late rounds, it becomes a mixed bag. You're drafting based on opportunity, breakout potential, or even just hoping for touchdowns.

Final Takeaways

If you’re looking for a clear, black-and-white takeaway from this study, here it is:
On average, a player on a team projected to win 8.5 games or more finishes as a positional top-12 (i.e., a “position_1”) about 65% of the time.

That consistency showed up across QBs, RBs, WRs, and TEs. But here’s the key, this data should not be used in isolation when you draft. It’s crucial to layer it with ADP. This entire study is built on results already priced in by ADP. I'm not suggesting you should reach for a WR with a 3rd-round ADP at pick 10 just because he's on a high-win team. That's not how this data is meant to be used.

That’s where the round-by-round breakdowns matter. ADP tends to be directionally correct. The real value of this data lies in how you break ties. For example, if you're torn between Player X on a 5.5-win team and Player Y on a 10.5-win team, and all other things are equal, the data supports leaning toward the player on the better team.

And that brings me to my final, simple takeaway, one that isn’t revolutionary but is still worth reinforcing:

In early rounds, it's generally safer to draft players on good teams.

These players tend to offer higher floors and often just as much ceiling, or more, because they benefit from better offensive environments. In later rounds, though, the logic flips a bit. Players on bad teams may have lower floors, but they can present real ceiling upside thanks to increased opportunity and less target competition. If you're willing to swing for the fences late, low-team players can pay off.

So overall:
Lean on strong teams early. Take calculated risks on weaker teams late. And always draft with context.

Page updated

Google Sites

Report abuse