Character Stats 101

[NaturalHazard]

can you please add: "CON IS NEVER A DUMP STAT" ?

can you devote an entire page to this line and make it as big as possible? thank you.

[tgun]

Why not just suggest that all characters should start 18 Con and put all lvl ups into Con....

Since that is exactly what most forum posters advise.
It is what the Uber players al claim.

I've never seen someone give this advice. Could you maybe find a couple examples of it?

[Sirea]

LOL, yeah paladins -- they need like every stat but they'll settle for dumping WIS most of the time.

Fixed that for you. Starting base WIS and equipping an item is enough to be able to cast all your spells. An AC Paladin especially will want INT for Combat expertise.

Why stop at 12(14 ) Con?
Why not just suggest that all characters should start 18 Con and put all lvl ups into Con. And Take Toughness feat and enhancements, and wear Grtr False life items, Minos Helms and GS HP items........MINIMUM!
And make that the official Turbine opinion.

Since that is exactly what most forum posters advise.
It is what the Uber players al claim.

What is your aversion to hit points?

Besides, everyone knows that a character that sacrifices all those build points to CON is only good for shieldblocking rats in the Vale

[OverlordOfRats]

I would suggest adding under Skills Vs Feat in the tip section:

Skills cannot be reset, choose wisely.
Feats may be reset once every three days, for a large sum of platinum (in-game currency) and a Dragonshard(available in game).

edited to include Kourier's and sirgog's comments.

[Kourier]

I would suggest adding under Skills Vs Feat in the tip section:

Skills cannot be reset, choose wisely.
Feats may be reset once every three days, for a small sum of platinum (in-game currency) and a token(available in game).

That's enhancements you're thinking of. Feats require a dragonshard, which becomes quite expensive at high levels.

Also, min-maxed players don't max out con. That's not a good return on investment. And while it's definitely possible to make a decent hp character with low con, a little build investment produces a big result - 40+ hit points (The equivalent of a shroud item).

[OverlordOfRats]

That's enhancements you're thinking of. Feats require a dragonshard, which becomes quite expensive at high levels.

I meant feats since it is mentioned that enchacments can be changed. Edited token to say Dragonshard. Thank you

[yawumpus]

This is sheer arrogance. What you are saying is that new players are too stupid to figure out things on their own if you give them the pluses and minuses. Instead they should just do what they are told to do because we know best.

That isn't Character Stats 101, that is here's the build play it.

If the purpose isn't pros and cons then the document has the wrong purpose. It is totally worthless without that element.

And plenty of people would argue that such is a good idea. On second though, just replace the currently approved "paths" with Tihocan's.

There is a reason for this. You could explain every rule in complete detail, but with first hand experience at all levels of play, you are not likely to come up with a build that will contribute to the party all through to cap (and even less to endgame and epic).

This isn't some opinion power gamers have, this is an often repeated experiment. Look up the "things I wish I knew" thread. CON IS NOT A DUMPSTAT is the most often repeated claim. DDO isn't pnp is repeated less often, but it is up there. 3.x pnp players know the srd forwards, backwards, and sideways. It is completely possible that they come to DDO with a better understanding of the rules than most DDO players (much more, since the DM has to handle the rules in front of them, not like turbines servers). Nevertheless, pnp players are notorious for weak builds that are painfully rerolled. This isn't just reading a quick guide to building characters, this is having the multiple rulebooks entirely memorized. Their builds will work fine in D&D, but this isn't D&D.

The other issue is that once a player builds his carefully thought out toon, he is dumped out on Korthos. He gets 20 bonus hit points, and gets another 10 "false" hit points when he runs the "storehouse secret" and picks the "right" choice (and equips it). At this point, a barbarian has 42 hit points from his level, the bonus, and the belt. Bumping CON to 14 (the powergamer minimum) would give him a bonus of 2, or a 5% increase. Our newbie is being a bit clumsy at the keyboard and gets hit a bunch, so he decides to increase his dexterity. He drops CON to 10 and pumps DEX to 14. While I have no idea what a Korthos mob hits on, it is far more than a 2. I'll guess a 10, but suspect they have a 0 BAB. His battleworn chain has a max dex bonus of 2, so AC is increased by 2. If the mobs hit on a 2, he would be hit just under 10% less often, but if it is 10 he would be getting hit 20% less often. So now he gets hit 20% less often for a price of 5% less hit points. This is perfect! Lets keep this character.

So while giving away 30 hit points to first level toons does a world of good to newbie survival (I'm from the pre-3.x give new toons max hit points - "ow, my hit point!" era), it teaches some very bad ideas on character building. That 5% penalty will become a 15% penalty and that 20% bonus will become a 0% bonus (much quicker for raging barbarians, but soon enough for all but "powergamer" AC builds). While not maxing strength is bad for a barbarian, imagine the poor wizard who didn't max his intelligence (most spells hit just as hard without for a good long time, especially if wearing the grotto robe), or the rogue who dumped strength (all too common), but maxed dexterity (who knows what happened to intelligence).

I will claim that shoving newbies into cookie cutter builds is a good idea. Cookie cutter builds work. The difference between a unique, usable build that will help the party complete the quest and an utter gimp that will ride around in a backpack (until replaced by some handy prayer beads) is not something that can be easily be determined by someone who hasn't played the game, even if they have the srd memorized.

[jkm]

personally, i'd suggest you do the following for the stat scores. take the calculators you have on the class pages and put them under each stat. most new players have zero understanding of how choices made at level 1 affect them at level 11+.

so for example, Con - blah blah blah min 12!!

See effects on CON score vs Damage Taken
[dropdown_charclass] [textbox_ConScore]

[add in level appropriate bonuses for each of 1,5, 10, 20 to False Life/Con Items etc]

Fighter, 14 CON

Level 1 - 10*1 (Levels) + 2*1 (Con Bonus) + 20 (Heroic Durability) = 32
Avg Damage for Quests (melee) = 2
Avg Damage for Quests (spells) = 3

This is a safe CON score for this level

Level 5 - blah blah blah

Level 10 - blah blah blah

Level 15 - blah blah blah

Level 20 - add in +6 for con stat item, +2 tome + 30 GFL = 20*10 (Levels) + 6*20 (Con) + 30 (GFL) + 20 (Heroic Durability) = 370
Avg Damage for Quests (melee) = 55
Avg Damage for Quests (spells) = 75

then put a warning up if they can be killed in 3 shots or less

You may want to consider more CON or the Toughness Feat

[wax_on_wax_off]

After some discussion with Tolero, we both agreed this was a good point! So I've added "Keeping this Ability Score no lower than 10 is highly encouraged, even if you're making a character who isn't a melee style." to the entry concerning CON.

No lower than 10? I kick people from my groups if I find out they have 10 starting constitution.

I think it is certainly true that after investing in your prime ability score for your class that 1-1 investments into constitution will give a greater return on investment towards survivability than any other options 99% of the time.

Additionally, survivability 99% of the time translates into overall effectiveness as when you have a good HP buffer you can play more aggressively which will lead to more skill/fun.

Obviously there are exceptions to a general "6 point investment into constitution" rule but they are few and far between.

[sirgog]

suggesting everyone keep there con above 12 .. funny ...

you do realize without telling them why you do little to no justice to them.

for each 2 points of con you get 1 hps per level ... so taking a con 8 you will have 40 less hps then a con 12,

if 40 hps is the difference between you suck and your uber .. then do it .. but dont take the suggestion that you NEED 12 con cause that is not entirely accurate and is highly pending on your class selection and multiclassing choices.

You can easily hit over 350 hps with the right gear with a 8 con at lvl 20. it just means getting the right gear.
If you feel you need more then 350 hps .. then simply add 20 hit points to 350 till you get to the number you want to have and figure out how many points to put in con. The more you put in con the less you can put into other stats so chose wisely.

350hp is a terrible number. It's just enough that you will be killed over and over again by taking two consecutive Fireball effects in epics. (DBF deals ~160-240 damage when Maximized, 130-210 after a resist; two of those will often hit within a space of 1 second and kill a 350hp player)

The other main consequence of dumping Con is that people myDDO you, see that you have a build utterly unsuited to getting through the quest/raid at hand without wasting an enormous amount of the healers' resources, and kick you from group. That's a pretty miserable experience to have happen to you several times starting at around level 14, and is likely a time at which people will ragequit the game.

Starting with a CON of 14 guarantees this will not happen. 12 is borderline.

IMO the game should force you to have 6 build points in Con on all toons until you have unlocked Veteran Status, at which point these kid gloves come off and you are allowed to make your own mistakes.



On saves, I'd mention that not all traps involve Reflex saves. A Horrid Wilting trap (there's some in the Shroud) involves a Fortitude save. An arrow trap makes an attack roll just like a monster.



Oh and on respeccing - I'd echo Overlord of Rats:

"I would suggest adding under Skills Vs Feat in the tip section:

Skills cannot be reset, choose wisely.
Feats may be reset once every three days, for a small sum of platinum (in-game currency) and a Dragonshard(available in game). "

except I'd change it to say 'a large sum in platinum' instead. For veterans, 80k PP is pocket change, but for newbies, feat respecs are expensive. It's not uncommon for me to need to get plat from an alt to fund a feat respec.

[wax_on_wax_off]

And, as Talon pointed out, that makes 14 CON the new average -- meaning anything less than 18 is really weakness.

14 Con IS the average (well, 6 point investment), at least among anyone with a clue, though it is a *median* value, not a mean value.

This means that the majority of occurrences of constitution is a 6 point investment. A majority of the exceptions will be those characters that have a greater investment (on arcanes or melees without anything better to spend it on). A very small group of builds might make an intelligent decision to invest less than 6 points (valiance 1.0 build for instance) while the rest will be gimp/noob/newb/flavour builds who don't know any better.

Additionally, in any situation where the extra HP from a high con *aren't* needed, you're better off first looking at skipping taking the toughness feat (what I did for my arcane). 1 extra feat > 4 build points usually.

[Zenako]

One of the problems I see is that many of the people posting about CON on this thread are focused on ELITE end game and EPIC level questing. On those settings, yes more HP are almost a requirement to survive outside a bodybag. HOWEVER, I would suggest that there is a rather large segment of the player base that never intends to play those types of quests. That building characters to survive in those quests, with full groups, will in fact gimp them for play in the fashion those players are more likely to engage in. Small groups or solo play.

So perhaps a little disclaimer section would be appropriate. For general play in solo and small groups (2 or 3) that certain basic stat levels are appropriate, but that for the current End Game and EPIC play (which often presumes 32 point builds as well!), that other conventions about stat allocation are customary.

A 28 point starter character has some serious tradeoffs to make in final stats if you drop a lot of those points into CON. It is not a clear win for many character types if solo/small group is your expected play mix.

[Zachski]

Saying that we should dump all other stats for constitution is a strawman. That's not what anyone is saying, merely that constitution is a necessary stat for all, but certainly not the ONLY stat that's important.

For one thing, 8/6 con builds can only be accomplished by really REALLY experienced players that likely already have tons of gear in the first place.

I.E., not newbies or even a vet rolling an alt on another server.

High HP != Zerger. A newbie that rolls, say, an Elven Wizard with 18 Int, 6 Con, and 18 Dex would find themselves probably avoiding a lot of damage early on due to AC.

Then at around level 10, they find that their high AC is becoming meaningless, enemies seem to flat-out ignore it, and casters murder them faster than they murder casters.

And then they find out that they really do need con, because there's some damage that just can't be avoided, and a party mistake that you can survive from is better than a party mistake that you die to. I mean, in a perfect party, sure, it's possible that Sorcerers and Wizards and all the other squishies don't get hit, but wouldn't you rather be able to survive mistakes than demand that every single person you meet is perfection incarnate?

So, what do they do? They get upset, rage that nothing indicated how important HP was (just telling players that CON raises HP does nothing in a scenario where they are used to games where casters have extremely low HP), get furious at Turbine for "exploiting players" by "demanding that they buy a Heart of Wood to correct a mistake they didn't know they made", and then quit the game, calling it stupid for having such a "sharp learning curve"

It is better that they are told to have at least 12 con. This still gives them plenty of stat points to put into other things. HP is both a necessity and a cushion, and the best way to avoid damage is to mitigate it through means like Blur and Displacement and flat out positioning yourself correctly. It's easier to learn if you have more HP than if you have less, but then rely on a flat stat like AC to save you from doom.

Remember, while an 8 CON build is possible, you need so many resources that newbies generally don't have. Let them learn when they're more experienced how to do it when they can handle it, rather than giving them a path that they are not ready for.

Also, remember that your build doesn't affect just you, it affects other players who party with you. After all, a healing-specced Cleric that has to raise you from the dead constantly is a drain on party resources. Not to mention you kill a 10% XP bonus for everyone if you die. And newbies with 8 CON will die more than newbies with 14 CON. No man is an island, especially in this game.

Which reminds me...

I think one thing you should add to the guide, FordyTwo, is that different classes have different HP growths, and thus different CON needs. A Barbarian has the highest HP growth, and so needs the least amount of CON to be on par. Fighter and Paladin have the second highest HP growths, so they actually need a bit more CON than Barbarian. Rogues, Rangers, Bards, Clerics, Favored Souls, and Monks have less HP growth, so they need more CON. Wizards and Sorcerers have the least HP growth, so they need more CON to compensate.

As such, I think you should add that, in fact, the less melee oriented you are, the more CON you need to compensate. If we're talking absolute bottom line, 12 for Fighters and Paladins. 14 for Rogues/Rangers/Bards/Clerics/Favored Souls/Monks. And 16 for Wizards and Sorcerers. Barbarians should go for higher CON regardless of their high HP growth so that they have the possibility of taking the PrE that lets them sacrifice HP to deal more damage with each swing.

Maybe at least add a note that says "Be careful about where you invest your ability points. Fixing them requires either starting your character completely over, or purchasing an item called a Lesser Heart of Wood from the DDO store, which may require you to purchase points using real money."

[Antheal]

* Intelligence

Learning ability and reasoning! Affects skill points gained at each level and the amount of bonus spell points for some classes.

* Wisdom

Willpower and intuition! Affects your willpower (Will save) and the amount of bonus spell points for some classes.

* Charisma

Personality and magnetism! Determines how well you influence non-player characters (NPCs) and affects the amount of bonus spell points for some classes.

Would it really be too much effort to state which classes?

[Therigar]

14 Con IS the average (well, 6 point investment), at least among anyone with a clue, though it is a *median* value, not a mean value.

Not average as a mathematical term.

When D&D first started there was the concept of the "normal" or "average" human being whose stats could be expressed in D&D terms. Initially 10 was the average. Anything above 10 was exponentially better and anything below was exponentially worse.

Adventurers were special because they had one or more stat in the exponentially better category and few if any in the exponentially worse. In the very early days when stats were rolled using 3d6 and in order this didn't always work out. But, the mathematical average of 3d6 was 10.5 which served to set the 10 point "average" or "normal" parameter.

As adventurers when we had a sub-normal stat it was often very problematic. We'd either work out an in-game rationale or we would, often, reroll. It made generating characters a full night's game session in many cases. It wasn't long before we went to 4d6 and discard the low number.

The notion of set minimum stats is relatively speaking new (although not unique to DDO having been used in other D&D products) as is the distribution of build points. The minimum stats can be thought of as representing the "average" or "normal" population that is not cut out for the adventuring life. Only adventurers can have stats above the "average."

What Talon is saying is that DDO's gaming community has effectively shifted this with respect to CON. The average isn't 8 it is 14. This influences the game because it puts a 3 HP per level boost into effect that monsters have to overcome in combat so monsters hit harder. It makes characters live more easily which means monster HP and/or AC gets boosted to keep them at an appropriate challenge level. All of this is w/o any intentional effort -- it is simply the byproduct of player bias and its effect on the game.

And, to some degree this is justified. After all, it takes an adventurer to deal with most monsters -- otherwise any old pole dancer would be doing it.

Keep in mind that this isn't about the pros or cons of CON per se. It is about how information gets presented to the new or begining player that is the target audience for this 101.

No matter how you phrase it, it is incorrect to say CON is never a dump stat (except in the most technical sense as was pointed out in an earlier post). What is correct is to note that not investing some build points into CON is something that should only be done for extraordinary reasons. Among those might be that you plan to solo almost exclusively or you are building with only 28 build points and have stat targets that cannot be met any other way.

Let's say you are building a rogue using 28 build points. You want to maximize AC because you know that AC ~70 is nearly untouchable thru Vale of Twilight and you have access to the monk class. You want Combat Expertise because of the +5 to AC and you want GTWF because you know that is key to melee combat. You prefer halfling because that is the uber race and besides Bilbo was a thief and that is the basis of all D&D halfling = thief lore. You also know that STR on halflings is low but it is a key element in all damage calculations.

Alright, it is now time to make some choices about your stats. You need INT for the CE, it is the key stat for DD & Search and it affects your skill points and you need a lot for rogue. You need DEX for the GTWF and for the impact on AC. You need WIS for the monk synergy plus it affects Spot. You need STR in order to not suck in melee and with a 6 base you'll be overburdened as soon as you grab the second or third really heavy item (or get hit by exhaustion). What would you like in those stats?

I'm guessing 14 INT. If you go 13 its enough but you lose a skill point, plus 14 is an even number which D&D and DDO rewards. You've got to have 17 DEX and you're new, so tomes are probably not in your thinking. But 16 will do even though 18 would be better. The even number here is even more important because race and class enhancements along with stat increases are an even number.

Then there's WIS which you can leave at 8 but then you're not really getting much out of that monk splash. Probably should raise it to something. Not really sure what yet. And of course you need points in STR. Hardly seems worth it stopping at 12. If you go to 16 that's a huge investment but 14 would be nice.

Now, I'm going to stop at 13 INT and 16 DEX with 14 STR -- just 3 of the stats I'm looking at right now. That is 5 + 6 + 10 build points. I've used 21 of 28 build points and not touched WIS or CON.

I need to know the pros and cons of choosing CON over WIS. If I put 6 points into CON I'm probably putting the last point into INT and leaving WIS at 8. It means I won't get to my ~70 AC goal. OTOH if I'm not even thinking about tomes I've got to toss 3 points into DEX to get the 17. Even more probable if I get forum advice to put my stat increases into STR (you know that's what you thought when you read I was putting them into DEX).

So, what to do with the last 4 points?

How realistic is the goal of high AC on a 28 point halfling rogue/monk? If I get there how survivable is the character if it has lower HP?

I recently calculated a beholder proof 77 AC for Therigar if I reincarnate him as a halfling and a minimally buffed 85. DEX target is 34 and WIS target is 28 including +2 tomes and +6 items. That means base plus enhancements of 26 and 20.

On a 28 point build with 19/1 split a character will hit 26 DEX with race & class enhancements plus stat levels in DEX. They'll have only a 14 WIS max. That is -4 from my beholder proof 77 number. Taking AC effects from class differences into account drops the numbers another 7 points.

It still gives a highly respectible 66 beholder proof AC which is almost guarantee against damage in Vale and below quests on all but natural 20s. Toss in a Dusk Heart and you've a character nearly untouchable.

So, is it worth leaving CON at 8 for that character?

Some people will say yes. Some will say no. But the key thing is that the individual player should be the one to decide. They need to know all the elements that go into the build and stat decisions before starting the character.

Thankfully, I'll never use DDO's 101 guide. As rolled out today it is not actually helpful and, after reading people's responses in this thread, it is clear that actually helping players make their own choices because they understand the game and the consequences of each choice isn't the goal.

The 101 will be like the default builds -- useless. A fine example of Turbine engineering.

[sirgog]

One of the problems I see is that many of the people posting about CON on this thread are focused on ELITE end game and EPIC level questing. On those settings, yes more HP are almost a requirement to survive outside a bodybag. HOWEVER, I would suggest that there is a rather large segment of the player base that never intends to play those types of quests. That building characters to survive in those quests, with full groups, will in fact gimp them for play in the fashion those players are more likely to engage in. Small groups or solo play.

So perhaps a little disclaimer section would be appropriate. For general play in solo and small groups (2 or 3) that certain basic stat levels are appropriate, but that for the current End Game and EPIC play (which often presumes 32 point builds as well!), that other conventions about stat allocation are customary.

A 28 point starter character has some serious tradeoffs to make in final stats if you drop a lot of those points into CON. It is not a clear win for many character types if solo/small group is your expected play mix.

It's not so much that dumped Con works in Normal and sucks in Elite.

It's more than dumped Con works just fine on all difficulties from levels 1 to 8 or 9 (long enough for you to get attached to the character). If you mostly run Normal and Casual, you might get to level 11 or so just fine. Then and only then does your early mistake start to ruin your character, even in Normal difficulty quests.

[Theolin]

Something about you only get the "next" bonus to X if you have an even number for your stat. Ending(lvl20) on a odd number doesn't help

[wax_on_wax_off]

Now, I'm going to stop at 13 INT and 16 DEX with 14 STR -- just 3 of the stats I'm looking at right now. That is 5 + 6 + 10 build points. I've used 21 of 28 build points and not touched WIS or CON.

Wow, a perfect storm.

Ok,

1. This assumes that someone owns the monk class but not 32 point builds which is unlikely.
2. This assumes that someone is trying to make an AC build without the resources to back it up, which is very newbish (resources implies high level character which implies 32 point builds, see point 1).
3. This assumes that someone is newbish enough to attempt to make a hugely MAD (multiple stat dependent) character while being very new to the game (see points 1 and 2).

Overall, I find your scenario implausible at best.

Additionally, you're trying to make this 1 issue responsible for multiple character creation challenges. Namely;
1. Don't make MAD characters until you have 32 pointers and have enough resources to provide +2 stat tomes at least.
2. Don't make an AC build until you know the pain involved.

[Artos_Fabril]

[...]
And, to some degree this is justified. After all, it takes an adventurer to deal with most monsters -- otherwise any old pole dancer would be doing it.

Keep in mind that this isn't about the pros or cons of CON per se. It is about how information gets presented to the new or begining player that is the target audience for this 101.

No matter how you phrase it, it is incorrect to say CON is never a dump stat (except in the most technical sense as was pointed out in an earlier post). What is correct is to note that not investing some build points into CON is something that should only be done for extraordinary reasons. Among those might be that you plan to solo almost exclusively or you are building with only 28 build points and have stat targets that cannot be met any other way.

Let's say you are building a rogue using 28 build points. You want to maximize AC because you know that AC ~70 is nearly untouchable thru Vale of Twilight and you have access to the monk class. You want Combat Expertise because of the +5 to AC and you want GTWF because you know that is key to melee combat. You prefer halfling because that is the uber race and besides Bilbo was a thief and that is the basis of all D&D halfling = thief lore. You also know that STR on halflings is low but it is a key element in all damage calculations.

Alright, it is now time to make some choices about your stats. You need INT for the CE, it is the key stat for DD & Search and it affects your skill points and you need a lot for rogue. You need DEX for the GTWF and for the impact on AC. You need WIS for the monk synergy plus it affects Spot. You need STR in order to not suck in melee and with a 6 base you'll be overburdened as soon as you grab the second or third really heavy item (or get hit by exhaustion). What would you like in those stats?

I'm guessing 14 INT. If you go 13 its enough but you lose a skill point, plus 14 is an even number which D&D and DDO rewards. You've got to have 17 DEX and you're new, so tomes are probably not in your thinking. But 16 will do even though 18 would be better. The even number here is even more important because race and class enhancements along with stat increases are an even number.

Then there's WIS which you can leave at 8 but then you're not really getting much out of that monk splash. Probably should raise it to something. Not really sure what yet. And of course you need points in STR. Hardly seems worth it stopping at 12. If you go to 16 that's a huge investment but 14 would be nice.

Now, I'm going to stop at 13 INT and 16 DEX with 14 STR -- just 3 of the stats I'm looking at right now. That is 5 + 6 + 10 build points. I've used 21 of 28 build points and not touched WIS or CON.

I need to know the pros and cons of choosing CON over WIS. If I put 6 points into CON I'm probably putting the last point into INT and leaving WIS at 8. It means I won't get to my ~70 AC goal. OTOH if I'm not even thinking about tomes I've got to toss 3 points into DEX to get the 17. Even more probable if I get forum advice to put my stat increases into STR (you know that's what you thought when you read I was putting them into DEX).

So, what to do with the last 4 points?

How realistic is the goal of high AC on a 28 point halfling rogue/monk? If I get there how survivable is the character if it has lower HP?

I recently calculated a beholder proof 77 AC for Therigar if I reincarnate him as a halfling and a minimally buffed 85. DEX target is 34 and WIS target is 28 including +2 tomes and +6 items. That means base plus enhancements of 26 and 20.

On a 28 point build with 19/1 split a character will hit 26 DEX with race & class enhancements plus stat levels in DEX. They'll have only a 14 WIS max. That is -4 from my beholder proof 77 number. Taking AC effects from class differences into account drops the numbers another 7 points.

It still gives a highly respectible 66 beholder proof AC which is almost guarantee against damage in Vale and below quests on all but natural 20s. Toss in a Dusk Heart and you've a character nearly untouchable.

So, is it worth leaving CON at 8 for that character?

Some people will say yes. Some will say no. But the key thing is that the individual player should be the one to decide. They need to know all the elements that go into the build and stat decisions before starting the character.
[...]

You make some good, well spoken points. In particular, I agree that the way to get new players to build better characters is to explain to them what decisions they're making at character creation, which are easily reversible and which require a re-roll (or LR) to fix. Letting them make decisions, even mistakes, is fine. Explaining the long-term repercussions of those decisions is better.

But I disagree with some of your assumptions:
1) A new player will know enough to build towards a 70 AC, but not enough to know that past vale it will still be virtually worthless, or pros and cons of HP (to speak nothing of "grazing hits", damage that bypasses AC, etc)

2) 16 dex for a halfling is 6 points, the 17th is only 2 more, not 3.

3) You would get recommendations to put level-ups in strength, rather than to not attempt an AC build rogue splash monk on 28 points as a first character.

4) For that build, after advising against it completely, I'd propose 13STR (power attack)/17 dex GTWF/14 con (HPs)/13 int (CE)/12 wis (AC)...

-about that advising against it completely: a rogue doesn't need all that much AC, the goal is to not have agro, if you do have agro, you're better also have the enemies blinded, so that you get your sneak attacks, if the enemies are blinded they've got a 50% miss chance already (effectively AC=monsterAB+10). If you're not getting sneak attacks, you might as well be playing a rogue splashed ranger or fighter.

-By the way, did you forget the significant gear dependency that 70 AC build has? It's not a new player build in any case.

In the end, I don't think that the compendium is going to reach enough people, or impart enough knowledge to make a significant difference, and for any specific build, the boards are (or at least were, but that's a whole other soap box) a better place to get build advice than this has any hope of being.

As far as this compendium entry goes, maybe a better entry would be:

Constitution: This Ability Score affects your hit points gained per level, your fortitude saves, and the availability of Monks' Earth Stance and Half-Elves' Barbarian Dilettante Feat. It is also the basis of the concentration skill, which is important to Monks and Spellcasters. Many players WILL discriminate against low-con or low HP characters when forming groups and raids. A score of 14 is therefore often recommended, especially for newer players.

**Red meant to denote hyperlinks to the appropriate entries.

I don't remember where on the boards I first heard this, but...
"It doesn't matter what else your character can do, when HP hits -10, all you can do is ride in a backpack."

[wax_on_wax_off]

*snip*

I don't remember where on the boards I first heard this, but...
"It doesn't matter what else your character can do, when HP hits -10, all you can do is ride in a backpack."

+1. I tried to say similar things but you were certainly more eloquent