Submission #6521: Lobsterzelda's GC NFL Street "NFL Challenge" in 1:14:20.35

Nintendo GameCube
NFL Challenge
(Submitted: NFL Challenge)
(Submitted: NFL Street (USA).iso USA)
Dolphin 5.0
802090
179.8266952145011
30425
Unknown
Submitted by Lobsterzelda on 9/28/2019 7:08 AM
Submission Comments

The Challenge:

Not satisfied with the level of respect that their scrappy street team commands from other teams, New Team decides to go out and defeat every single NFL team. However, as if this challenge wasn’t daunting enough, they decide to go one step further. They won’t just beat each team – they’ll clobber every team 38-0, with every play against the opposing team resulting in a score. Finally, they will defeat an all star team consisting of NFL legends from years past, and prove that they are by far the greatest football team of all time!

The Game:

NFL Street was a game released by EA Sports in 2004 for the Gamecube. In the single player mode of the game (which is the main mode of game play), you are given an initial team of 7 players. Your goal is to complete challenges to gain development points, which allow you to improve the skill rankings of each of your players. After you become skilled enough, you can start taking on NFL teams and beating them. Each game is played with "street rules", which means there are no kickers, no penalties, you can tackle players into brick walls, and there's plenty of styling/taunting. In order to complete NFL Street Challenge Mode (the name of single player mode), you have to beat every NFL Team, (which all have their 2003 rosters) a team of all-stars from each division, and a game against the NFL Legends, which contains the best players in NFL history. Each game is played with a "first team to reach 36 points wins" ruleset. Once you have spent enough time practicing and building up development points, you will be able to just barely defeat the NFL Legends in a tightly contested championship game…

The TAS:

…or you could simply obliterate every team beyond all logic or reason, which is what this TAS does.
In NFL Street, the play that the computer will call on the next down is decided based on where the computer’s players were on screen when the last play ended, how long the last play lasted, and what your players were doing when the last play ended. By ending the previous play on a slightly different frame each time, you can keep trying out different options until you find a play where the computer quickly pitches the ball. From there, it’s a simple matter of calling an all out blitz in order to steal the ball and get a touchdown. By repeating this pattern on every single play, I am able to decimate each team with extreme speed. Furthermore, by ending each play on specific frames, I can skip the post-touchdown cutscenes that often play when you score a touchdown, which saves about 2 seconds for each cutscene I skip in the run.

The Dream:

When I was 12, I saw a TAS of The Legend of Zelda Ocarina of Time, and was blown away by what I saw. The level of precision and speed that was possible from TASing a game and pushing it to its limits was something I had never seen before. After watching through a few more TAS videos, I stumbled upon Adelikat’s TAS of the football game Tecmo Bowl for the NES. Seeing the crazy 30+ second long plays, lightning fast dodges, and across-the-field-and-back plays made me wonder what other sports games could have seemingly impossible tasks accomplished via TAS tools.
More specifically, I was obsessed with playing NFL Street at this time, and was curious as to how fast the TAS for this game was. I excitedly looked up “NFL Street TAS” online, but much to my disappointment, I soon discovered that nobody had TASed the game before.
So much time in NFL Street is spent just wracking up development points to make your players more skilled before actually playing any NFL teams. I knew that it should be theoretically possible to beat every NFL team without getting any development points in a TAS, but since there was no video of this in existence, I had no way to definitively prove that this was possible.
I hoped that someday an NFL Street TAS would be made. As I learned more about the techniques used to make TASes, I could picture what the end result of an NFL Street TAS would look like in my head: no development points being used, every play on offense resulting in a quick touchdown, every 2 point conversion try succeeding, every defensive play that the computer calls being a quick pitch play that gets caught/recovered in the endzone for a touchdown, and every post-touchdown cutscene being skipped.
However, the idea of actually TASing a game seemed to me like a daunting task that I wasn’t ready to learn how to do just yet. As such, there was little else I could do except wait and check once every year to see if somebody had finally created an NFL Street TAS.
Fast forward 10 years later to 2019. I found myself feeling bored one day, and decided to check if an NFL Street TAS had been made yet. Once again, a search still turned up nothing. Suddenly, I thought to myself “If nobody else will do it, then why can’t I make a TAS of this game myself?”
As a result, I took on the challenge of learning how to TAS, and soon found ways to implement the ideas that I had had a decade earlier. Eventually, I finished my work, submitted it to TASVideos, and had it published on the site.
Over time, I began to notice how the quality of the TAS fell short of what I had envisioned for the games' TAS as a kid. Cutscenes played after about half of all touchdown celebrations in my TAS, not all 2 point conversion tries were successful, some plays took extremely long, and some plays by the computer didn’t involve pitching the ball. With newfound skill and expertise in hand, I decided to set out to make a second TAS which would implement all of the things that my first TAS had been lacking.

The Process:

As I experimented more with the game, I found new techniques to save time throughout the game. Post touchdown cutscenes could be manipulated to not happen by delaying the end of the previous play by a few frames, the next play could be more easily manipulated to be a pitch play by holding different directions while throwing and catching the ball, the 2 point conversion plays themselves could be altered by changing what frame the previous touchdown was scored on, and the start of a game could be delayed by a few frames so that the CPU would call a play which would result in a quick and easy touchdown for me.
I spent the last month working on this TAS, and what you see here is the result. All of the goals mentioned above are met in this TAS, and it obsoletes my earlier movie by about 10 minutes!
The CPU has approximately a 1 in 25 chance of calling a pitch play on any given play. This means that there is a 1/25^164th power chance of getting 164 pitch plays in a row (which this TAS does). Assuming that there are approximately 10^80 protons, neutrons, and electrons in the observable universe, that would mean that if every sub atomic particle represented one possible combination of 164 plays chosen by the computer according to the probability above, then you would have to search through about 2*10^29 different universes before finding one particle that got all 164 pitch plays in a row. The power of TASing to manipulate luck to make the seemingly impossible become overwhelmingly likely shines through here. However, given that a person playing NFL Street in real time can’t use frame advance or save states, it’s unlikely that the RTA world record for NFL Street will come anywhere close to the time achieved in this TAS. Considering that the current RTA WR for the game is 5 hours and 40 minutes and the astronomically low probabilities described above, this seems like a fair assumption for me to make.

The Bottom Line:

With all of that out of the way, my dream is now realized. Furthermore, New Team finally achieves its goal of winning every game in a blowout and forcing the CPU to always pitch the ball, proving that free will is an illusion that hides reality from humans that New Team is by far the greatest football team to ever walk the streets.
Admittedly, a lot of the entertainment of the previous run was sacrificed to save time for this TAS, but since the last run was accepted to the vault anyways, my main goal for this TAS was to beat the game as fast as possible regardless of entertainment sacrifices ¯\_(ツ)_/¯.
I’ve said this with a few of my submissions, but it’s especially true for this TAS: This is the TAS that I have put the most work into of any TAS I have ever worked on. I hope you enjoy the fruits of my labor!

Masterjun: Judging.
Masterjun: Accepted to the Vault as an improvement to the previous movie.
feos: Pub.
Last Edited by adelikat on 11/1/2023 7:58 PM
Page History Latest diff List referrers