The excitement surrounding the NFL’s annual schedule release has died down. We now know not only the New York Giants will be playing in 2025, but when they’ll be playing them.
The Giants’ schedule is daunting on paper, and we immediately began speculating on how the 2025 season will play out. Ed took the optimistic path and is predicting a 7-10 record for the Giants. Cynthia Frelund of NFL.com is modeling an average of 5.4 wins for the Giants.
Almost a year ago, I decided to take the human element out of it and used Pro Football Network’s playoff prediction machine to simulate the season 100 times.
That actually turned out to be an interesting exercise and it’s prediction of a 4-13 season for the Giants was disappointing but ultimately accurate. I don’t think their machine could have predicted Daniel Jones being released and Dexter Lawrence suffering a season-ending injury, but it was accurate.
So I decided to run the experiment again and see what the results were. Once again, the results are interesting — though we’ll have to wait and see just how accurate they are.
How many games will the Giants win?
This is the question on everyone’s mind. Last year, the simulations predicted a 5-12 season for the Giants (4.77 wins on average), with 4-13 as the most common result.
This year, the simulations predict a similar 5-12 record, but with an interesting twist. The average wins is actually a bit worse, at 4.67 wins, however the most common result was a 6-11 season (22 instances).
Las Vegas set the Giants’ over/under at 5.5 wins, and this seems to fall in line with that.
All told, there were a combined 30 instances of three- and four-win seasons (11 and 19, respectively), compared to a combined 33 instances of six- and seven-win season (22 and 11, respectively).
It’s interesting that, at least according to the simulations, the Giants are as likely to come away from 2025 with a seven-win season as they are to finish with three wins. And while the simulations predict that it’s more likely that the Giants will finish with one win (4%) than eight, nine, or 10 wins, they do give the Giants a slightly better chance of having “eight or more” wins (5%).
The Giants’ lone playoff appearance came with a 10-7 record on the third simulation.
Rank in the NFC East
Given how the simulations panned out, it shouldn’t surprise anyone that the Giants finished fourth in the NFC East far more often than not.
There were 71 instances of a fourth-place finish, as opposed to 27 instances of a third-place finish.
The Giants did finish second in the NFC East twice, once making the playoffs and once just missing them. Ultimately, the Giants’ rank in the division — and their final record as a whole — usually came down to their season series with the Dallas Cowboys. The Cowboys frequently had a very similar record to the Giants, and in the 29 times that the Giants finished in third (or second), it was Dallas who was in the basement.
Draft picks
For the most part, the simulations are predicting that the Giants will draft highly again in 2026.
Of the 100 seasons I simulated, the Giants ended up with a Top 5 pick in 53 of those seasons.
That, of course, didn’t necessarily spell disaster. It was, frankly, odd that the Giants were as bad as they were in 2024 and didn’t get the first overall pick. So the inverse can be true as well.
For instance, in the 33rd season I simulated the Giants finished with a 6-11 record, were third in the NFC East, and had a -31 point differential (losing by an average of 1.8 points). However, they also but also held the fifth overall pick. The record might not thrill many fans, but the peripherals suggest that the team actually played well, took a lot of playoff caliber teams right down to the wire, and still got a valuable draft pick.
However, as you can see, the first overall pick was the most common result of the simulations, with all of the one-win seasons and most of the two- and three-win seasons resulting in the top pick.
The slight bump at 14th overall was due to the number of 8-9 and 7-10 seasons I got. The 27 total fifth, seventh, and eighth overall selections were split between the five- and six-win seasons, with the exception of two four-win seasons.
Looking ahead, the Giants might be in the market for an offensive tackle, wide receiver, or cornerback in next year’s draft. If so, picking in the rand of fifth to eighth could be a good spot.
Takeaways
That there were 21 instances of five wins suggests a slightly pessimistic attitude toward the over-under. That’s likely influenced by the Giants’ high strength of schedule, as well as their lackluster (to say the least) performance last year.
Analytically speaking, past offensive performance tends to be more predictive of future success than defensive performance. But rather than lending clarity, that actually adds a layer of murkiness to the predictions. The quarterback position is the biggest reason for that year-to-year stability and predictive power on offense. And, as we’re well aware, that’s also the area of biggest change for the Giants on offense.
Frankly, we don’t know what the Giants’ offense will look like with Russell Wilson at the helm.
As it was last year, Wilson was more efficient and effective than Daniel Jones. Wilson had a 38.4 EPA and 2.8 points added over replacement, compared to 34.5 EPA and 3.6 points below replacement.
So if Wilson plays to the level he did a year ago for the Steelers, he does represent an improvement at the most important position on the field. Granted, it’s a modest improvement, but an improvement nonetheless.
It’s also notable that Wilson suffered a calf injury early in training camp, which limited his ability to work with the Steelers’ starters during some crucial phases of the build-up to the season. A full off-season and training camp as the clear starter could yield a better season than Wilson had a year ago, and make for a more significant improvement on offense.
If you’re looking for a more concrete reason for optimism, I kindly direct you to the defensive side of the ball.
Last year, the Giants had a final point differential of -142 points on the year, or an average of -8.4 points per game.
The average through 100 simulations was an average point differential of -39.86 points per season, or -2.3 points per game. It would mark an incredible improvement for the team to go from losing by an average of more than a touchdown to less than a field goal.
For reference, that was about the Atlanta Falcons’ point deficit last year. They finished with an 8-9 record.
Without getting under the hood of PFN’s algorithm, I can’t tell for certain on what they’re basing their projections. If they’re basing their projections for the Giants on last year’s performance, the (potential) upgrade at the quarterback position as well as health along the offensive line could put the Giants’ offensive output over expected. There was roughly a 5-point difference between Daniel Jones and Russell Wilson last year. If the defense can keep opponents within a field goal, that difference could lead to the Giants outperforming projections.
Of course, it’s also possible that they’re accounting for an improvement in offensive output and that 5 points is factored into the narrowing point deficit.
In my takeaways from the Giants’ 2025 schedule, I wrote that a six-win season is probably a reasonable projection. However, I also suspect that the Giants could have an outsized influence on the playoffs and be a team that nobody wants to play in the final month of the year. If their defense is forcing tight games and their offense is putting up points as well, the Giants could have the look of a team ready to break out. Even if the final record might not reflect it.
But that’s why we have to play the games to find out for real.