np: Anti-Hax - It's your lucky day...

...or is it? Not on Smogon's Shoddy server it isn't. Unless you've been living under a shamrock for the past two weeks, you've noticed one of the most heated threads in Policy Review history, which revolves around, well, luck in competitive pokemon. Before X-Act posted actually this thread, he, Doug and I spoke on a conference call for a little over half an hour about his findings with statistics and his confidence in being able to arrive at a formula that would determine to what extent luck was responsible for a player's win. We've launched a secret hax-free server to test this formula and work out the kinks, and we're finally ready to see just how ready we are to part ways with luck in competitive pokemon battling.

We will be implementing the hax clause on the ladder tonight after the close of ratings. As with the most recent and future Suspect Tests, Doug and I are going to keep the formula from the public so as to prevent anyone from being able to game the system. Rest assured that the three of us—who average around 33 years in age and have been working with math on a professional level for longer than many of you have even been playing pokemon—have the smarts and experience to be able to initiate such a complicated movement in pokemon.

Just remember that this may be for the benefit of competitive pokemon, an ideal consistent with every Suspect Test. Now that, after weeks of actual testing, we have arrived at the perfect formula, we are trying to determine whether a hax-free game is in the best interests of competitive pokemon.

As usual, this thread is to be almost completely devoid of theorymon. This thread will be closed until Doug gets the new ladder up, at which point he'll open it.

I apologize for not getting the Anti-Hax Ladder up and running last night like we planned. I worked on it for hours last night, and encountered several unexpected technical problems with the Smogon University server. There's no problem with the Anti-Hax Clause itself, but adding a new ladder to SU has been much more difficult than I initially expected. I am working to resolve the problems today. I will continue to keep the community apprised of my progress in this thread.

I would like to clarify something that Jumpman did not cover in his OP:

Although the overall community response to the Anti-Hax Clause has been extremely positive, many people have expressed frustration over the mis-perception that the formula behind the clause is "completely secret".

The formula is NOT completely secret.

The detailed math behind the formula has been kept secret, but the functional elements of the formula have been described and discussed at length in this PR thread. In that thread, we have laid out all the key metrics that are used for measuring luck and determining the appropriate winner of a battle. Unless you have a PHD in Math, that information is all you realistically need to know about the formula. If you are a PHD in Math, and you have sufficient expertise with competitive pokemon, and you have active involvement with the Smogon community and policies -- then you are already on the team that developed the formula in the first place.

The only reason anyone else would need to know the exact formula, is if they plan to specifically tailor their battle strategy to take advantage of the formula. That is not the purpose of the formula. It is not a new game mechanic, in and of itself. And, I really don't want to hear any pedantic complaints about the definition of "game mechanic".

Pedantic Whiner said:

"If this formula is used to determine the winner of a battle, then it is certainly a game mechanic!"

Click to expand...

Yeah, yeah -- I've heard it. Congratulations on your ability to split hairs. I'm not buying it, so please stop bitching about it.

This formula is meant to measure a battle after it is complete. It is not meant to be a factor in the battle itself. For now, the only way we can ensure that, is if we keep the exact details of the math secret. In that way, no one can exploit the formula with specific battle techniques. The details of the formula will be given out on a need-to-know basis. And most people really don't need to know.

I have good news and bad news...

Bad news: I am still having problems with the new ladder on the Smogon Shoddy server

Good news: The Anti-Hax Clause is up and running on the Create-A-Pokemon server in unrated play.

We are internally discussing how to handle the problems with the new ladder on Smogon, and I expect we will have resolution in a few hours. I will post here as soon as the new clause is open for public testing on SU.

In the meantime, for any of you bursting at the seams to check out this exciting new development -- just log on to the CAP server and you can get a preview!

The Anti-Hax Clause is now live on the Smogon University server!

After much deliberation, we have decided to test the new Anti-Hax Clause on the Standard Ladder on the Smogon server. I realize this may come as a surprise to many of you, but there are several reasons for this decision:


Feel free to log onto the server, and see what we've been working on for the past several weeks. This thread is now open for feedback and discussion.

Excellent, can't wait to try this out.

At first I was skeptical but after playing a few matches I suppose it's not as bad as I thought it would be.

The homosexuality of this is beyond me.

Sceptile used Energy Ball.
It's not very effective...
Toxicroak lost 37% of its health.
Chloroform's Toxicroak fainted.
The battle has ended.
Analyzing battle data for luck factors...
Chloroform wins!
EeveeTrainer: WTF
EeveeTrainer: OMG
Chloroform: XD

Analyzing battle data for luck factors...

Click to expand...

i love this thing

I believe that this is a bad idea because it ruins the chances of Togekiss and Jirachi being OU. It is a bad idea to change the metagame this much because now you don't want hax and Stone edge is a bad idea.

Redundant :p

Paranoid212, it doesn't punish the person who gets more luck. It punishes the player who played worse. Playing better and getting "hax" are not mutually exclusive; you can hax someone and still play better than them.

People can still use Jirachi and Togekiss without any real effect to the game, as long as they play better.

darkie said:

Paranoid212, it doesn't punish the person who gets more luck. It punishes the player who played worse. Playing better and getting "hax" are not mutually exclusive; you can hax someone and still play better than them.

People can still use Jirachi and Togekiss without any real effect to the game, as long as they play better.

Click to expand...

So, for example if I predict a switch and use a move that would normally have OHKO'd the switchin, and it misses, that gives me plus points, but if I flinch someone ten times in a row with a scarf Jirachi, it doesn't punish me?

If so, I'm all for this thing.

a crit or a para every once in a while is not going to throw the game. and there are other uses for jirachi and togekiss too, they are not OU for the sole purpose they can cause hax, they are both good pokemon even without the use of fliches and FP this new clause will just mean they are going to be used in different ways.

darkie said:

Paranoid212, it doesn't punish the person who gets more luck. It punishes the player who played worse. Playing better and getting "hax" are not mutually exclusive; you can hax someone and still play better than them.

People can still use Jirachi and Togekiss without any real effect to the game, as long as they play better.

Click to expand...

I think the main thing about this is that we don't know how the formula works with less-accurate moves; if you Body Slam with Jirachi and it paralyzes, for example, is that hax or not? What if it doesn't? Does it simply not count in the formula? What about Hypnosis?
Sorry if this is answered in the PR thread, but I think it only got a passing mention.

Paranoid212 said:

I believe that this is a bad idea because it ruins the chances of Togekiss and Jirachi being OU. It is a bad idea to change the metagame this much because now you don't want hax and Stone edge is a bad idea.

Redundant :p

Click to expand...

It doesn't work like that. Think about it - Jirachi and Togekiss flinching 60% of the time isn't hax, it's expected. That is the norm, and in an average battle, Jirachi will flinch an average of 60% of the time, and as will Togekiss. If they flinch 95% of the time, or only 15% of the time, then that can be considered as hax, and the formula will take it into account (I think). They are no more 'haxxy' than any other Pokemon.

This deacreases use in Stone Edge, Jirachi, Togekiss, and more popular choices. THESE ARE CORE ELEMENTS TO OU GAMING!! =(

And what about crits that dont matter? Like if he has 2 HP and I crit with a KO move anyway?

I question how this thread can conceivably be "free of theorymon" when no one actually knows the precise details of the formula besides the three of you. How do you suggest we adequately judge if the formula is working properly or not when we don't actually know what it's intended to do? If I have a battle overturned in my favour when I feel I still should have lost, is that a sign that the formula is broken, a sign that the formula has some minor inconsistency that only cropped up that once, or a sign that my definition of "too much hax" is separate from yours?

That last point raises another question: where are you deciding what hax is too much and what isn't? Even without the mathematical details, there is no denying that luck IS an integral part of Pokemon, or we wouldn't have the option to choose between a weaker 100 accuracy move (Flamethrower) for reliabily killing a lot of things but missing a few crucial kills, and a stronger, less accurate move (Fire Blast) that will get all the kills you need but without perfect reliability. I know you claim to have the perfect formula, but while I don't doubt you have a formula that perfectly decides whether there was too much hax based on YOUR view of how much hax is okay, I can't help but wonder how in line your view is with the majority of players. Granted it's an extremely difficult question to answer, but an important one to think about.

And also, is this thread meant to be solely discussing if the formula works properly or not, or can it be used to discuss if the formula should be used at all? If the former, will the latter be opened to public (ie non-PR) debate at any point?

This isn't pokemon anymore. This is smogonmon.
Seriously, hax is part of the game. Yes I hate it, but you have to learn to deal with it.

I think I'll make a team based around paralysis with Togekiss, Jirachi and Machamp hmmm.

I doubt it does but can this formula assess how crippling luck is? I realize that it takes the damage that the attack would have otherwise done and weighs that but CHing a Pokemon that plays a bigger role in stopping the foe's entire team is worth much more than scoring a CH on just an average Pokemon.

How is this is a better and more fair/accurate solution than simply removing all added effects from the game? If they are going to just be nullified later, why not just totally drop flinches, forget about CHs, remove added effects, fix accuracy to 100%, etc. and have the first one who causes all 6 of the foe's Pokemon to faint win? It would remove any confusion in who the winner is when it's all said and done (please don't say that it modifies the way the cartridge because a "hax clause" is just as guilty of doing so).

Either way, I can't help but feel as if the "luck management" aspect of Pokemon is being severely neglected here. I'm sorry to say it but if you're looking for a perfectly competitive game, then Pokemon isn't for you.

alive said:

At first I was skeptical but after playing a few matches I suppose it's not as bad as I thought it would be.

Click to expand...

Agreeing with this, it isn't that bad. I'd like to know more about how the system works but I'm sure you guys coded it very well, so it looks pretty nice.

Can't wait to test this out more!


Edit: Lol a few battles later >__>

bon jovi the purpose of a ladder is to measure the skill of the person playing it. you are still playing pokemon, but you smogon rating will be different depending on how heavily hax played in the battle. as far as i can tell this will not be in place when it comes to tours and tournaments, but can anyone confirm?

Please keep all comments in this thread on-topic. We don't need a bunch of stupid remarks or quips. If you want to contribute to the thread, then do so intelligently.

I like Stall Teams the best right now, just watching them and playing with them. Basically, if you can construct a stall team that simply stays alive and you let your opponents rack up crits while you spike and heal, you can pretty much win with the hax factor shifting the balance to your favor because you don't attack or crit while they are. If the opposing team does not crit you and your team goal is to stay alive and strive, they will just have a tough time stopping you.

DDMence, SD Lucario, and SD Scizor are also having field days so far from a few battles witnessed, basically if they are capable of sweeping a team but fail to do so because the damage is minimum it really gives that analyzing hax factor a swing in their favor.

Alright I'm not really sure if this clause is working correctly :/ I just had a battle against Wakalord and we both agreed I didn't deserve the win. I got, I believe, two crits, numerous para-haxes (Celebi vs Celebi...), and I attacked through confusion a few times too - the crits specifically were really game changing. After analyzing the data I was still given the win. What gives?

tbh too I don't really think I played that well either, I was just toying around!

edit ok now I had a fucking joke of a battle with darkie and I didn't get the win.

be sure to check the log for all crits that occured, any misses / instances of hax, as well as damage outputs. it could be you got minimum damage a lot and he got maximum damage everytime. or you could have missed an attack like fire blast early game or something?