1-50 version 3 (my ideas on an upgraded 1-50)

edit:
If you guys just read the bold and the intro you will get the gist of what I am saying. Gosh darn lazy kids :stuck_out_tongue:

intro:

So, this has been inadvertently in the works for months, just kind of bouncing around my noggin whenever the topic came up. It is far from fool-proof and is more of an attempt to sprout some ideas. I have thought up, along with conversing with other members of the forum, some slight alterations to the 1-50 that will please far more people than did the H3 system.

Now 343 is probably putting the finishing touches on H4’s ranking system, but if enough people like a different idea maybe 343 could make some changes to what they are currently using. And if worse comes to worst maybe it can be something looked into for H5.

The major benefits are:
-Giving any level of remotely competitive players who are looking to test their skills, and be rewarded for it, a place to do it.
-Providing key “Halo-esque” elements to H4 to recapture fans turned off by Reach’s major changes.
-Providing much more instantaneous gratification that an Arena type system (necessary to keep players interested).
-Drastically reducing the desire to make new accounts.
-Drastically reducing the effectiveness of boosting and deranking.

Now, most of these benefits are simply provided by having a 1-50 in itself, but there will be 2 major changes:

1. Reduce the effects a team mate’s low trueskill and high sigma has upon your rank after continuously playing with them.

Okay this is pretty difficult to explain (especially when I don’t exactly know the terminology or how exactly trueskill works) but I will do my best.

(Background on trueskill - please read even if you already know how trueskill works, and let me know if I am incorrect about anything)

To my understanding, a new account with have the lowest trueskill and lowest sigma (sigma being “how set in stone” your rank is). Due to your low sigma, if you win a game, you will be drastically thrown up in trueskill in an attempt to more accurately match you with players. When you lose a game, your sigma will increase drastically, in an attempt to correctly put you at the rank you belong, along with reducing your trueskill drastically as well. I believe as you win games your sigma will increase as well, although I am pretty sure it is not as much as a loss effects it. Sigma is the reason people become “rank locked”; If you win a game, then lose a game, then win 2 games, then lose 2, for hundreds of games, your sigma will continue to go up but your rank will not. True skill will believe you are exactly where you belong and your sigma will become so high that it would take 50 wins in a row to get you to the next level. If you did finally make it to that next level, your sigma would be drastically reduced, and it will take less games (if you were winning every game) to get to the next rank, and then less than that for the next rank, etc.

(Background on boosting)
Now sigma also plays a major factor in boosting, when you a play with a buddy who has a very high sigma, they have a much greater effect on your trueskill after a win than would someone with a lower sigma. Also, when you play with someone of a lower rank, and beat players of a higher rank, because trueskill intended for you to lose that match, you will go up drastically if you win, and only go down slightly when you lose. inversely, when you play a team who trueskill expects you to beat and you do win, you barely go up, but if you lose, you will go down a lot.

So, if you have a team mate with a very high sigma, along with a low trueskill (leading the system to think you will lose everygame), it will “boost” you drastically. This is why and how boosting and deranking occured.

(solution)
My, what I believe to be, simple solution to illegitimate boosts of your trueskill, is simply to cut the “boost” that you get in half everytime you play with that same player. So, if you are playing as a random, you will still get the full “boost” if you carry someone with high sigma and low trueskill, but it is when you party up with them and play with them again that you only get half the boost, and the next game half of that until it is reduced to that of any other player (whatever the value 343 determines). After 5 or 6 games of playing with a “booster account”, it no longer has any benefit over a regular account, this making boosting extremely difficult to do. Also there should be a slight decay rate on the “Ban” from receiving trueskill boost from that player so that if you play the same random guy again a few days later in MM, you still get that full boost.

The decay rate needs to be enough to keep players from being able to easily swap out booster accounts, but with minimal detriment to legitimate gameplay.

I would think this would have minimal repercussions, as anyone who is legitimately playing with a friend who is high in sigma but low in trueskill is either looking to help their friend out, in which this change has no negative effects upon, or just wanting to play with them, in which the HS/LT (High sigma, low trueskill) account with just act like a more “normal” account in the effects that it has upon the other player.

2. Next is The part which Benefits the most competitive players, reduces a 50’s desire to make a new account, and adds a new element to any players who enjoy ranked gameplay

Simply add an arena style % within each rank. Pretty much as if there were 50 divisions instead of 5 and you can’t lose your rank from not playing, only your %. To reiterate your rank is the same way as halo 3 and Is not affected by other players going up or down from that rank. Just that within that rank you will have a % telling you how high within that rank you are compared to others who are also that rank. This will be the most useful for 50s who are still looking for competition and something to strive for as once they attain their 50 they can either leave it and just stay a 95% 50 or they could strive for the 1% 50. There may or may not need to be need to be seasons for the % part, I haven’t really gotten that far.

  1. in regards to the difficulty of attaining a rank, I haven’t really put much thought into it. We want it be hard but not impossible or we could have similar problems that H2 had. I think maybe 5-10% harder than H3 would suffice if the % thing in itself doesn’t solve the problem.

conclusionthingy:

In regards to everything else how the EXP system or cR system or whatever will work, I haven’t really thought about that.

So what do you guys think? My opinion is obviously going to be very bias towards the 50’s, but I am attempting to please everyone with the idea, so let me know what you think. I think this is the best of both worlds taking what Arena had to offer but without sacrificing all that the 1-50 had to offer.

edit 9/6/12:

See this post for counter argument on “people will just maintain multiple boosters and in will making deranking even more rampant1”

Ouch , I wasn’t looking forward to reading that many words…

I just want my double exp weekends back!

you can read mostly the bold and it will make sense. and half of it explains how and why boosting occurs, so if you already know (or don’t care) you can just skip the “Background on trueskill” and “background on boosting” paragraphs.

Cmon guys -_- don’t be scared, just read the intro and the bold.

I support your idea, however percentages on each level seems a bit much? I mean 1% 50 would be great and all but wouldn’t it take forever to find a game? Perhaps if the percentages just showed how close you are to the next level I would agree, so that people can see directly how much each win affected their ranking.

I agree with this and would be pumped if it was in Halo 4. There is a NEED for a 1-50 or skill based rank system in Halo 4.

Only reason your post isn’t getting that many replies is because it is entirely too logical. By reducing the boosting effect in the way you just suggested (granted I only read the bold and intro, got work I need to be doing) all negatives about 1-50 are largely accounted for and dealt with.

Hopefully 343 sees this and all the other threads supporting a 1-50 comeback!

Sounds absolutely amazing to be completely honest. Would allow people to play with friends more freely with each other while drastically reducing the boosting abilities. Also, full incentive to play again even as a 50 would really mean a whole hell of a lot to me.

Me and my friend at one point created new accounts together nd played doubles only with eachother for something like 150 games without a loss, we were like level 32 XD

Would your sigma suggestion work both ways, so that when i (with a near perfect win/loss ;D ) played with some of my worse friends, they would actually be able to rank up every once in a while? (this would kind of take away 1/2 of the reasoning for sigma because in theory i could boost my friends up past there true level, however not ANYWHERE near as effective as in halo 3 with my 18 game booster :stuck_out_tongue: )

Thanks again for another thought out and intelligent post. Really starting to like you :stuck_out_tongue:

> I support your idea, however percentages on each level seems a bit much? I mean 1% 50 would be great and all but wouldn’t it take forever to find a game? Perhaps if the percentages just showed how close you are to the next level I would agree, so that people can see directly how much each win affected their ranking.

I don’t think it would have to be strict on matching by percentages.

Perhaps for 40-50 you get the percentage? Honestly though, its an amazing idea… because there is no incentive to play once your a 50, and this would definitely be an incentive for me… a HUGE one. It would be very cool to have a percentage to level as well :slight_smile:

I like your idea

Instead of just cutting the boost effect in half or whatever you said, why not just make it so that your team mates have no influence on your rank. With your system, people will just make multiple boosters and switch off playing, while it will also punish legitimate players who play together.

Also, making ranks so that you are the only one who can see them will get rid of a lot of incentive to cheat.

I think your system could work, but would it not just be easier to set limiting factors on maximum value of sigma.

This would reduce the effect of boosting and allow for players who perform above average consistently but have a high sigma to not get ‘level-locked’ allowing for a more dynamic ranking system.

From the bold, red and blue text that I read I agree

> Instead of just cutting the boost effect in half or whatever you said, why not just make it so that your team mates have no influence on your rank. With your system, people will just make multiple boosters and switch off playing, while it will also punish legitimate players who play together.
>
> Also, making ranks so that you are the only one who can see them will get rid of a lot of incentive to cheat.

Honestly, I feel like a lot of the incentive comes from a visible rank. Call me a show off, but I would have nowhere near as much incentive without it. I would still prefer it over reach’s system or no rank at all, but it just wouldn’t be the same.

It wouldn’t “punish” people who play together. It could in theory help them. Two people playing together with amazing records makes them never level up, the opposite effect of boosting. This would allow for it to stop basing the games you win with your partner on each others truskill, meaning you would rank up as if you were playing without them.

Yes, I thought about having multiple boosters too. With rank being slightly harder to obtain and the method mentioned, you would need around 15 boosters to get someone a 50 in any reasonable amount of games.

Say you have 15 booster accounts.

You play with each of them 2 games and get 1 1/2 games worth of truskill boost. The best boost in all of halo was 18 games, average was anywhere from 20-30+ 25/1.5= 16.6.

Creating a booster takes ALOT of time. If you had to maintain 15 booster accounts, you could probably only boost 1 account per week. It would be more effective for a person to just play by themselves.

I’m curious, what’s the point of sigma?
Why aren’t our ranks equally ‘set in stone’ for every rank, and every game?

> I’m curious, what’s the point of sigma?
> Why aren’t our ranks equally ‘set in stone’ for every rank, and every game?

MU

MU was an innovative way of making the 1-50 more fair. If you lost with a team of people that had HORRIBLE win loss you wouldn’t move down to far. If you played with a team of people who won every game you wouldn’t go up to far. It was a nice try, however exploited massively by creating accounts with 1-10 win loss ratio (horrible MU), and then pairing it up with a new account (no sigma). This would trick the system into thinking that this new account was massively carrying someone who lost almost all his games, to victory. Making you level up super fast.

However, this new system would reduce the negative effects of the old system and make it near flawless in my opinion.

Sigma

Sigma was basically just to keep players who got a couple lucky games from instantly ranking up if they’ve been a level 44 for 300 games, and to allow new accounts to move up fairly quickly instead of having to play 200 games to get to level 30.

At least that’s the best i can make of it.

> Instead of just cutting the boost effect in half or whatever you said, why not just make it so that your team mates have no influence on your rank. With your system, people will just make multiple boosters and switch off playing, while it will also punish legitimate players who play together.
>
> Also, making ranks so that you are the only one who can see them will get rid of a lot of incentive to cheat.

This. By trying to stop people playing with boosters constantly, you’re eliminating people playing with teams or their friends. And knowing me, I’ll end up removing everybody off of my friends list once I play with them a ton just so I can rank up better.

I mean, i appreciate your idea and all, but the game has already been made, even if this was full-proof and everyone supported it and it got 343’s attention… it’s too late.

> > Instead of just cutting the boost effect in half or whatever you said, why not just make it so that your team mates have no influence on your rank. With your system, people will just make multiple boosters and switch off playing, while it will also punish legitimate players who play together.
> >
> > Also, making ranks so that you are the only one who can see them will get rid of a lot of incentive to cheat.
>
> This. By trying to stop people playing with boosters constantly, you’re eliminating people playing with teams or their friends. And knowing me, I’ll end up removing everybody off of my friends list once I play with them a ton just so I can rank up better.

It wouldn’t “punish” people who play together. It could in theory help them. Two people playing together with amazing records (truskill) makes them never level up, the opposite effect of boosting. This would allow for it to stop basing the games you win with your partner on each others truskill, meaning you would rank up as if you were playing without them.

> > Instead of just cutting the boost effect in half or whatever you said, why not just make it so that your team mates have no influence on your rank. With your system, people will just make multiple boosters and switch off playing, while it will also punish legitimate players who play together.
> >
> > Also, making ranks so that you are the only one who can see them will get rid of a lot of incentive to cheat.
>
> Honestly, I feel like a lot of the incentive comes from a visible rank. Call me a show off, but I would have nowhere near as much incentive without it. I would still prefer it over reach’s system or no rank at all, but it just wouldn’t be the same.
>
> It wouldn’t “punish” people who play together. It could in theory help them. Two people playing together with amazing records makes them never level up, the opposite effect of boosting. This would allow for it to stop basing the games you win with your partner on each others truskill, meaning you would rank up as if you were playing without them.
>
> Yes, I thought about having multiple boosters too. With rank being slightly harder to obtain and the method mentioned, you would need around 15 boosters to get someone a 50 in any reasonable amount of games.
>
>
> Say you have 15 booster accounts.
>
> You play with each of them 2 games and get 1 1/2 games worth of truskill boost. The best boost in all of halo was 18 games, average was anywhere from 20-30+ 25/1.5= 16.6.
>
> Creating a booster takes ALOT of time. If you had to maintain 15 booster accounts, you could probably only boost 1 account per week. It would be more effective for a person to just play by themselves.

People will maintain 15 boosters though. That is 15 times the deranking.

Reducing the boosting effect is a work around to the problem. Getting rid of the affect a players sigma and mu have on you ranking up is a fix to the problem. Boosting is eliminated, except for the only legitimate kind: going in with a player who is a higher level than you when you are on a new account.

> Cmon guys -_- don’t be scared, just read the intro and the bold.

I think %ages only on 50s would be nice, you could know if your a bad 50 or a good 50.

And i also like all ur ideas.

I know this may sound like a troll question…but what is up with the sheer fascination that people have with numbers?

In part I ask this because at it’s base the 5 divisions are more concise if you look at it 0-10 (Iron) 11-20 (Bronze) 21-30 (Silver) 31-40 (Gold) and 41-50 (Onyx). Granted I know at least going back to the days of H3 the discrepancy of the range could be cause for concern of some of those that go for a higher level of play but the percentiles in the Arena were made to correct that.

It really seems that the Arena would’ve been better off with a higher pop (Because Arena is the only playlist set that appears to have the tighter Trueskill in place),a more visually displayed rank, and obviously the settings that the community had wanted. (Some say Seasons hurt Arena too, but I’ve always seen them in the vein of how sport rankings are always in flux and made it seem like a more desirable feature on a competitive front. [Though season roll overs could be handled better in the future if the Arena system stayed to show some recognition])