Anyone watch Deadliest Warrior on Spike?

[charlie and algernon]

a couple computer geeks pit different warriors that have been seperated either by different time or location. they test their different weapons and then enter all the data into a computer to determine the winner out of 1000 battles.

Of the ones I've seen the following warriors have won their battles:

Iroquis over Gladiator
Pirate over Knight
NY mob over Japanese Mob
Russian Spetznez over American Green Barrett

The episode now is pitting China's Shaolin Monks against the New Zealand Maori Warriors.

[HarukaTheTrophyWife]

1. It's a good show

[charlie and algernon]

2. yeah, I like it

the only problem I see is that in group battles, it just ends up coming down to the special hand to hand weapons. Why not just have those two face off to begin with if that's all that's going to matter.

[Godhumor]

8. I believe they script a battle with just the winner related to the computer sim

i.e. most of the simulation numbers actually show the short range weapons rarely get used, but the scripted for TV battles need to show off all the weapons featured.

[deucemagnet]

3. I'm watching.

It looks like the Shaolin weaponry is going to make the difference in this episode.

[charlie and algernon]

4. yeah, they're pretty bad ass

that last guy just took out a guys eyes, then continued to turn a head into swiss cheese with the eyes still on the knives. :wtf: :puke:

[deucemagnet]

5. Next week: William Wallace vs. Shaka Zulu.

Fucking brilliant. :headbang:

[charlie and algernon]

7. yeah, that looks awsome

though i'd almost rather they just pit Wallace's Scottish army vs. Zulu's army. That way they could get into comparing their tactics and strategy.

[jobycom]

6. I did that in third grade with comic book characters, but no one thought it was cool.

I was so way ahead of my time.

[Godhumor]

9. The methodology of the show is pure shit but I love watching it for...

The weapon demonstrations and usually really well put together fight scenes.

That said, as someone who runs simulations for a living, the way they decide winners is a pure joke.

[charlie and algernon]

13. what would be the better way to determining the winner?

[Godhumor]

14. Well, 1000 simulations is small potatoes

Especially as they have had multiple shows where wins have been decided by under 10% of the vote. 10,000 itinerations at minimum--or, at the very least, if a win is decided by under 100 kills they rerun it the base 1000 times and do it best 2 out of 3. Essentially, there's too much random flux in that small a sample based on random simulation.

Spoilers

Take last week. The result was decided by "7 kills outside the standard deviation". That was used to justify declaring a winner when the results were essentially tied. His statement is nonsense--one standard deviation only accounts for about 2/3 results in normalized testing. Honestly, when you use random numbers for testing you keep running simulations for larger and larger test runs until the results are consistently in one direction or another.

/end spoilers

Obviously I don't think the show should show them rerunning the simulations, but I would like to see "After x runs at 10,000 trials, the monk consistently wins about 70% of the time."

On a less numbers wonky note, the squad battles that feature 5 on 5 fights for some reason post results as if it were a 1 on 1 battle (i.e. total kills add up to 1000 instead of the 5000 to 9000 that should happen.). Just a bit odd.

[Broken_Hero]

10. Its a good show, just watched

both the ones they showed tonight the Beret vs. Spetnatz, and Maori vs. Shaolin....also, it was actually the Apache vs. Gladiator. :D

[Broken_Hero]

11. There was one more, the Viking vs. Samurai

I've watched like the first half three times, but always got sidetracked with other issues, and never found out who got the nod.

[charlie and algernon]

12. haha, I saw that one too, but never saw who won either

ended up having to run out right before the final battle.

[LostInAnomie]

15. It's an interesting show, but their methodology has HUGE flaws.

My biggest problem is though, is that dead is dead. If your major weapon causes a huge wound that would cause someone to bleed out in seconds, and the next person's weapon causes slightly more damage, there might be statistical difference but there is no practical difference. Dead is dead.

I had a big problem with the Spartan vs Ninja episode. Not the result, mind you, but the method used in gathering their stats. When testing the weapons they used martial arts experts who have been using the Ninjas' weaponry for years. Therefore, they are incredibly skilled and proficient with the weapons (probably more so than actual ninjas). For the Spartans though, they used a "weapons expert" (who the hell knows what that means) and had him using weapons that he more than likely hadn't been practicing with for years (like a shield). Obviously, a real Spartan would be VERY familiar and VERY skilled with all the weapons and would be able to use them far more effectively than the "weapons expert".