So, I rolled 5 18s

[Endur]

A character such as this needs a great name.

Conan or Aragorn or Groo, or some other name that conveys a destiny of things to come.

As a role player, you will have the burden of striving to play a perfect specimen of humanity with a fearsome destiny.

 

[Glyfair]

Conan or Aragorn or Groo, or some other name that conveys a destiny of things to come.

No, a Groo needs at least one horrible stat to put in Int (and likely at least two)

 

[Ed_Laprade]

Now, that doesn't mean it didn't happen. It just means the odds against it are stacked rather higher than 50/50.

I know that the odds aren't really 50/50. The point, which everyone seems to be missing, is that there are a lot of people in the world who roll dice for various reasons. That being the case, it's simply a matter of time before someone does it. And in this particular case, it happened to be the OP. Just at a guess, I'd say it probably happens once a year or so somewhere in the world. (And considering how many people there are in the world who might be rolling dice for whatever reason, it could be as high as once per day!) But unless it happens to an RPGer we aren't likely to hear about it. To anyone else, except maybe a statistician, it's just a curiosity.

 

[Ed_Laprade]

No, a Groo needs at least one horrible stat to put in Int (and likely at least two)

Definitely at least two, INT and WIS. Probably three: CHA as well.

 

[Glyfair]

Just at a guess, I'd say it probably happens once a year or so somewhere in the world

Doing an estimate (based on very loosel numbers) of the odds of someone rolling 5 straight 18's on 4d6 take the best 3 this is what I get. If about 32 million people do this every second for a year, you can expect it to happen about once. (And no one really rolls 5 stats in 5 seconds)

Now, this isn't accurate for this case, because there is the wildcard of the 6th stat. I'm sure someone with a more mathematical background can break it down further and find out the odds of actually doing so. However, I really don't think even 3 million people are actually doing this at any one time, and certainly not over an entire year. I doubt the odds shift too much further than that.

 

[Infiniti2000]

There ARE racial max's.
...
Maybe "racial normals" explains things better?

Maybe rereading the stat explanations in the front of the PH will help explain things better.

No, there are no racial maximums. There are starting maximums. There's a huge difference. While you're rereading that chapter, please find the section that specifies the racial maximums for, say, dwarf, as opposed to any other race.

Even then, there's no explicit maximum. There's only an implicit maximum given a particular stat-generation method (the rules actually say that the modifier is between -5 and +5, but that is provably an error--just consider Int). For any number of d6 keeping the best 3d6 you get an 18 as the maximum. For point-buy, get a number limited buy the number of points and the DM's cost structure above 18 (note, no rules on that either). But with "select whatever you want" there's no limit unless you impose a limit. So, let's pretend you're the DM and tell me the limit.

I ask again (rephrased), you'd have no problem with me just choosing stats at that limit, right?

 

[Erekose]

This reminds me of the "perfect bridge deal" stories. This is when every player is dealt all the cards of a single suit. The odds are 2,235,197,406,895,366,368,301,559,999 to 1. Yet, almost every year during the height of bridge (the mid-20th century) there would be a newspaper story about that being dealt. That many bridge deals weren't dealt. In fact, I'm not sure it's possible that many have been dealt ever.

Analysis by the experts traces it to three things. The first is the obvious, lying. A hand comes close and they want a story so they report it as a perfect bridge deal (helped by the fact that newspapers would actually print these stories without any evidence beyond the players anecdotes).

The second is the imperfect shuffle. People rarely actually shuffle cards randomly. In fact, when computer dealt hands started taking hold there were a lot of complaints about "freaky hands." Analysis showed that the hands were within statistical expectations. The problem was that people were used to the imperfect shuffles which tended to lead to flat hands.

The last is also obvious, cheating. A good magician can duplicate this deal many ways very easily among non-experts in sleight of hand.

I think these examples here are very good possiblities for most stories of straight 18 rolls. Plus, there are always oddities. However, when things are that far out of line of expectations, there is probably something else going on.

I could be being really stupid but you know I just have this funny feeling that this could be misinterpreted as someone accusing me of lying.

I know what I saw and I have described it without ambiguity - the fact that it is unlikely (perhaps very unlikely) does not mean it's untrue.

 

[Glyfair]

I could be being really stupid but you know I just have this funny feeling that this could be misinterpreted as someone accusing me of lying.

I know what I saw and I have described it without ambiguity - the fact that it is unlikely (perhaps very unlikely) does not mean it's untrue.

Not any specific case. However, with a number of "I saw someone do this" that are that far beyond expectations, you can assume that some of them (indeed, most of them) were because of variations on above cases.

Also, the above cases don't all deal with lying. The first clearly does. The last is effectively the same. The middle case deals with imperfect dice (we all know stories of dice that don't really roll randomly), dice rolling methods or something similiar (for example, bad randomization programming if you use a computer program).

 

[Erekose]

Not any specific case. However, with a number of "I saw someone do this" that are that far beyond expectations, you can assume that some of them (indeed, most of them) were because of variations on above cases.

Also, the above cases don't all deal with lying. The first clearly does. The last is effectively the same. The middle case deals with imperfect dice (we all know stories of dice that don't really roll randomly), dice rolling methods or something similiar (for example, bad randomization programming if you use a computer program).

Ah I see you are a master of making a simple post incomprehensible - I suppose I should have surmised this from the original response.

 

[Voadam]

I played in a game (run by Shark) where my character had I think 3 18s, 2 17s and a 16. as a 1st level character (actually ECL 2 with +1 LA aasimar race). I think his method was roll 6d6 for each stat take the best 3 rolls and do this six times for each stat. He changed the method a little later for lower rolls but I chose to keep mine for my noble fencer concept rogue

I've played in both high and low stat games and I enjoy them both.

When I used rolling character generation in the past as a DM I allowed different official methods (including human rolls from UA and the demi human ones from dragon) but required the rolls be done in front of me at the moment with no second chances. I currently use point buy methods.

It depends on the DM, what he will accept. Some say assign your own stats, others say if you made him fairly I trust you, others require rolling in front of them. If your DM accepts him and you want to play a high stat character then by all means talk to your DM about it.