Part 1: an in-depth discussion of the state of my team, how I let it get that way, why I decided to rebuild, what my plan was for the rebuild, and the first steps I took along the way.

Part 2: reaffirming the vision for my rebuild and detailing the next steps, along with a treatise on how I trade. The philosophy and strategy behind why and how I trade along with a discussion on how can I be a high-volume trader in leagues that are often derided as “inactive”. Also, a representative and complete set of trade negotiations from start to finish.

Part 2.5: Written after the season, I reflect on some of the thornier strategic questions I faced, detail the final roster-shaping trades I made before the deadline, and evaluate all of my moves both with and without the benefit of hindsight.

Part 3: Offseason negotiations and completed trades on through the rookie draft and some musings on where my team stood entering 2021.

Part 4: My “last look” at the rebuild and final evaluation of the process.

I said in Part 4 that I had enough data to evaluate the processes and I wasn’t super concerned with the outcomes, so that would be my final word. So why am I adding one more entry? Partly because I wound up putting up one of the best seasons in league history and taking home a title. And partly because I’ve been thinking about making this into a series at Footballguys, so I wanted to get in one last data dump so I could start editing down and distilling the key points.

As I mentioned, my team wound up taking home a title in spectacular fashion.

My final All-Play record was 148-39, or 79.1%. This is my new personal best across my 24 league-seasons in dynasty (15 in my “home” league, 9 in this staff league). (I’d averaged 74.1% all-play over the previous nine seasons in my home league, but that’s because 8 of my last 9 seasons finished between 73.5% and 78.6%.)

I acquired the last piece for my team (Austin Ekeler) after Week 5; my All-Play record since then was 104-17, or 86.0%. Finally, over the last seven weeks of the season my All-Play record was 75-2 (97.4%), which was the best by any team over any 7-week stretch in this league’s history (previous record: 72-5, 93.5%).

My points per game average (159.8) was the 3rd-best in league history. The average since the Ekeler trade would be a new league best at 162.7 (previous record: 162.3) and my closing stretch also set the record for points per game over any seven week span at 178.3ppg (previous record: 171.6).

My team featured the #2 and 10 quarterbacks, the #1, 2, 5, and 8 running backs, the #5, 14, 18, and 20 receivers, and the #1 and 11 tight ends. Nine non-quarterbacks scored 300 points this season and I rostered four of them. Seventeen non-quarterbacks scored 250 points and I rostered five of them.

All of this was especially sweet because last year’s rebuild graded as my “second-worst” season ever, ahead only of the inaugural 2013 campaign where I traded all of my productive players for injured stars and future picks.

Now let’s talk about how that happened. (Mostly it was incredible luck.)

• When I started my rebuild, my highest-value asset was Alvin Kamara. First I tried to trade him for Clyde Edwards-Helaire, but that wasn’t happening. Then I tried to trade him in a package for J.K. Dobbins, but the numbers weren’t there. So finally I settled for trading him in a package for Jonathan Taylor.
• In my biggest trade, I decided to throw in Deshaun Watson for Kyler Murray just because. Both were young enough to suit my rebuild and they were comparable values, but I felt like shaking things up. Murray has been a regular starter for me, while Watson hasn’t played a snap this year.
• Late in 2020, I wanted to trade Chris Godwin for Tee Higgins straight up, but it wasn’t happening. Instead, I traded Godwin for Miles Sanders and Miles Sanders for Tee Higgins. Because of the way the trades shook out, this also resulted in me trading O.J. Howard and a 2nd (wound up 2.12) for a 1st (wound up 1.09). And then Jaylen Waddle surprisingly fell to me with that pick. Waddle was the #14 fantasy WR, a regular starter, and a long-term cornerstone asset that I got essentially for “free” because I was unable to trade Godwin for Higgins straight up.
• Speaking of Tee Higgins, my Week 16 playoff game was against the team who I acquired Higgins from. Higgins put up 43 points. I had the highest weekly score, he had the 2nd-highest weekly score. If I didn’t have Higgins, I would have lost. If I didn’t have Higgins and he did, I would have gotten my doors blown off.
• I tried repeatedly to sell Leonard Fournette last year but could never get what I felt was fair value. Fournette was one of two players I held through the rebuild. He was the #5 fantasy RB this year despite missing two games so far. I started him seven times and he averaged 22 points per game in those seven weeks.
• The other player I held through the rebuild was Justin Herbert. I wasn’t seriously trying to move him last year, but I did entertain offers and was mildly tempted by a couple (most notably, Justin Herbert for Darius Slayton). Herbert was the #2 fantasy quarterback this year and carried me through long stretches with Kyler Murray out.
• I only completed one trade where the other party was the instigator. A leaguemate offered me Mark Andrews for the 1.07. At the time, it looked like a very fair trade overall, though I was excited to do it because tight end was my biggest weakness. Andrews was not only the #1 tight end on the season (by a significant margin), because of the insane TE bonuses he was the 2nd-highest scoring player at any position (including quarterback), trailing only Cooper Kupp. Again, that wasn’t a trade I sought out, it was offered to me unsolicited.
• The above points are especially relevant because I only earned a playoff bye because of tiebreakers. Had I lost one more game, I would have been playing in Week 15, which was one of the few weeks where I didn’t have the top weekly score. Instead, it belonged to one of my potential opponents (depending on how many extra losses I faced and who they came against).
• I earned the #2 pick in the rookie draft this year and selected Ja’Marr Chase. Had I lost one fewer game, I would have drafted 1st and taken Kyle Pitts. Had an opponent lost one more game, I’d have drafted 3rd and taken Najee Harris. Had I won one more game, I’d have drafted 4th and missed out on all of the “Big 3” entirely.
• Pitts and Harris are both fantastic players and I’d be thrilled to have either, but I don’t know if I would have earned a bye with either, and neither were as big of an asset in the championship week. Additionally, if I’d drafted Najee Harris it’s less likely I ever would have acquired Aaron Jones or Austin Ekeler and more likely I would have sold Leonard Fournette before he broke out.

(Which underscores the importance of evaluating based on processes instead of on outcomes.)

I’ve mentioned a few times that this would traditionally be considered an “inactive” league. Since the beginning of August 2020, the league has seen 20 trades. I’ve been involved in 11 of them (with 8 different teams).

A number of those trades were penny-ante, though (Antonio Gandy-Golden for Drew Brees; Pittsburgh’s Defense for Devin Asiasi and a future 5th; etc.) It has seen 11 of what I’d call “serious” trades (with the package on each side valued as high as a future 1st or more). I’ve been involved in 8 of them (with 7 different teams). There have been five trades that saw at least six pieces change hands and I made four of them. There were eleven trades with at least four combined players and picks involved and I was part of seven of them.

“Number of players and picks changing hands” is a poor metric for many reasons, but overall 59 players and picks changed hands in trades I was involved in and 29 players and picks changed hands in trades I was not involved in, giving me a 67.0% market share. (If trades were distributed randomly, each GM would be expected to account for 16.7% of the movement; proof is left to the reader.)

Out of the eleven possible trade partners, I made deals with 8 of them and *significant* deals (involving at least a 1st round pick’s worth of assets on each side) with 7 of them. Of the three GMs I didn’t trade with, I had productive conversations with one that nearly yielded a deal but came up just short (I wanted to trade Akers and Jonnu Smith for the 1.01 to take Kyle Pitts but he got cold feet and then Akers got injured). I had pleasant conversations with another that resulted in him updating his rankings because it turned out he agreed with me on player values (this is when I was well above market on Calvin Ridley). And the third… just didn’t have any players that fit my rebuild, so I never approached him. (But he’s a great guy and I’m sure I could have easily gotten something done otherwise.)

By most measures, this is an inactive league… for anyone but me. But I had no problem at all making small deals, making medium deals, making big deals. This is a good lesson– just because a lot of deals don’t happen doesn’t mean there aren’t a lot of deals there for the making. And if you’re having trouble getting things done, often the problem is not the league itself, but your approach to trading.

Perhaps getting deals done is my superpower. But if so, it’s a fairly mundane superpower and one I feel confident I could pass on to any interested party.

There are lots of ways to estimate the total value of a dynasty roster. Some of them I think are better than others. But provided the method correlates with actual dynasty value (however weakly), it is a mathematical certainty that increasing value over a long enough timeline will inevitably indicate an improving roster, which will inevitably result in more wins and more championship shares.

(The more strongly your preferred measure correlates with the actual underlying value, the shorter the timeline necessary for continued improvement to be meaningful.)

I like using Dan Hindery’s trade value charts at Footballguys because (a) his process strongly resembles my own, so his player values are significantly closer to mine than anyone else’s in the industry and (b) his values are explicit, fine-grained, and comparable across timescales (compared to something like dynasty ADP, where it’s hard to tell if the #6 pick in one month is worth more than the #6 pick in another, and also hard to tell whether the #2 and #27 player are more valuable than the #8 and #15 player).

One interesting way to measure dynasty team strength is to sum up the values of every rostered player in the entire league and divide by 12 to estimate how strong an “average” team is, then sum up the values of every player on your own team and compare against that baseline. In my experience, dynasty rosters tend to skew top-heavy so the median team usually comes in around 95% of average. (As a thought experiment, it’s possible for a team to have 200% of the average value, but virtually impossible for a team to have 0% of the average value, reinforcing that leagues should tend to skew top-heavy.)

According to Hindery’s value charts in August of 2020, my team was pretty middle-of-the-pack, checking in at around 95%. Mostly due to Fournette getting cut by the Jaguars, that fell to 92.2% in September, meaning I had a noticeably below-average team. Not only that, but my own spreadsheets showed that my players were aging and what value my team had was heavily loaded into 2020.

Basically, I had a team that was bottom-half and leaking value. Which is why I decided to rebuild.

If I had left my team as-is, it would have done exactly that; leaked a ton of talent and not won a championship. Per Hindery’s April trade value charts, if I’d played out the season without making any of my trades (but still making my free agent drop/adds), my roster would have fallen to 75.7% of league average, easily a bottom-3 squad, but still old and lacking the high rookie picks necessary to refresh. Again, this is just to reinforce that rebuilding was absolutely the proper decision in hindsight.

Instead, my roster post-rebuild graded out at 130.1%, making it one of the Top 3 teams in the league. And after winning a championship in 2021, my team is up to 163.6%, by far the best in the league.

But even in the “unlucky break” scenario (where instead of Murray, Taylor, Chase, Waddle, and Andrews, I wound up with Watson, Edwards-Helaire, Diontae Johnson, Rashod Bateman, and Rondale Moore) I’d have graded out around 120%, good enough for a Top 3 roster and still plenty young to keep a viable window open for years to come. Which means even independent of the specific outcomes, the process was sound.

To close, here’s one last look at my team before and after the rebuild.