Improvement table awarding a given value of improvements.

[garion]

I know you want to discount all MM leagues etc... but they still give us some interesting information. Lets look at a few specific positions from fumbbl and the difference in skill selection taken from the skill selection survey. While it is still an MM CR based match making environment it still appears to me as though there were more skills taken that we currently see.

this is LRB4 info- http://fumbbl.com/skills.php

Yes the basic skills are still the most common but that is not ever going to change, but after your Block Guard and Mb on bash teams or after the blodge SS of elves things look very interesting, also the doubles taken show far more diversity that we don't seem to be getting from CRP obiously we need the numbers but I would be very suprised if it showed as much diversity. Wouldn't you?

Lets look at an amazon catcher -
Double rolls - (48%) Nerves of Steel (24%) Jump Up (15%) Guard (3%) Dump Off (3%) Dauntless

So as you can see, because of many contributing factors, NoS was alot more popular as was Jump Up, Guard 3rd is not something you will see now. In CRP I would expect this to read - Guard, Mighty Blow, Pilling On for doubles, I doubt we would get past that.

Amazon Thrower (76%) Strong Arm (11%) Nerves of Steel (5%) Guard (4%) Pro (1%) Side Step

76% strong arm is a huge, again I would be very suprised in the current edition if we saw strong arm leading the way here, it will probably be high, but I would expect Guard to be first doubles choice again, so it seems as though probably becuse of spp costing rather than Tv costing that passing players were more common then rather than silly leader caddies, this is also a common theme running between all throwers if you check. Also Leader seems quite a rare choice back then.

Chaos Beastman doubles -
(32%) Claw (18%) Big Hand (14%) Razor Sharp Claws (11%) Foul Appearance (4%) Dodge

Obviously most of these skills are single rolls now, but the fact Foul Appearance got picked over Dodge is still very interesting now after its uber nerf I know we wouldn't even see it in the top 5 normal skill rolls. all we would see would be Mighty Blow, Claw, Pilling On, Frenzy, block,


Bull Centaur (37%) Stand Firm (35%) Dodge (18%) Frenzy (4%) Pro (2%) Catch

again first double Stand Firm, we know this will never be the case now, but it is certainly more interesting than the ubiquity of dodge. Again Pro and Catch as later skills? This will never happen now.

Thrower (67%) Strong Arm (21%) Nerves of Steel (3%) Guard (3%) Pro (2%) Jump Up
again strong arm is the most common and NoS still comes before guard.

Dwarf Runner (58%) Dodge (12%) Pro (10%) Strong Arm (6%) Nerves of Steel (3%) Sure Feet

Pro as second most common double. NoS gets a look in too.

high elf Lion Warrior (now called catcher ) (48%) Nerves of Steel (22%) Jump Up (11%) Guard (5%) Dump Off (4%) Dauntless

again, NoS is the most common doubles pick and Jump Up is picked over Guard. No MB or Frenzy even made it into the list.

Gutter Runner (53%) Very Long Legs (22%) Dauntless (5%) Horns (5%) Foul Appearance (3%) Pro

VLL most common double. Thats never going to be the case now, also FA and Pro get a look in, this would never happen now.

[dode74]

MM leagues have a very different dynamic to "normal" leagues or tournaments. The imperative is for TV-efficiency (so you can win in the MM environment), and those where there is a tournament (such as the FUMBBL cup, which I am enjoying watching when I can) often use the MM part (e.g. ) to team build for the tournament itself, making the imperative survival and player building rather than winning. Those dynamics will have a HUGE impact on the skills taken, which is why I am loathe to look at them. I am not discounting them, I am saying that the information they provide is based on an environment for which the rules were not optimised, and that therefore leagues must be our first port of call. If the issue is still there in leagues then we can say that there is a problem; if the issue is not there in leagues but is there in MM then the problem is the environment and some sort of house rule (perhaps the OP's one, or a variant of it) may be in order. (underlined because I think it's a really important point!)

The info you've given is great, but we need to look at how the setup has changed before we can draw any conclusions.
- Your Amazon Catcher, for example, wouldn't take Jump Up on doubles any more as it's normal access. Guard is still very high up the list. Your assertion that MB/PO would be the next two doubles, with precisely no information to back that up, belies your own prejudices more than anything.
- Amazon Thrower - again you make a bunch of assumptions.
- Beastman - FA was nerfed, as you say, making it less attractive. Claw is still right up there, as is RSC (which, as a combo, is EXTREMELY powerful). I think your top 5 misses tackle out, but the point is still that they require a LOT of skills to be as good as they can be.
- Bull Centaur - Old SF was very nice as a S skill, and dodge does part of what old SF did (help reduce turnovers) and can be used ot the same effect with BT, giving the BC back his all-important mobility.
- Thrower - NoS is a single now.
- HE catchers now make really good blitzers on a team which has only two.
- GR - VLL was changed dramatically iirc. FA was nerfed.

My point is that it may be the changes to the skills themselves which is a part of the issue. Some of the numbers you gave (runner with 58% dodge, Zon thrower with 76% SA) indicate that there was a lack of variation even in LRB3.

One of the things you don't mention is where there are similarities:
CW most common normal skills are Block, Guard, MB, Tackle, PO with Claw and RSC as a double. Now that is a killer.
Dwarf Blitzer common normal skills are Guard, MB, Tackle
Human Blitzers - Guard, tackle, MB with dodge on a double.

Sure, some things are different, but some are the same. My point is that until we can have some info on how skills are distributed in CRP then we have no comparison and anything said about what we think the CRP numbers might be is pure speculation based on our own biased perceptions.

[Darkson]

Jumping the gun a touch here
We've not even got as far as establishing what the real issue is, let alone a solution.

Just because you can't see an issue, doesn't mean it's not there.

[garion]

replies in red

MM leagues have a very different dynamic to "normal" leagues or tournaments. (snip)...... Yes I get this but if we see more variation even in the crazy world of MM it is still supporting this to a point

The info you've given is great, but we need to look at how the setup has changed before we can draw any conclusions. Yup
- Your Amazon Catcher, for example, wouldn't take Jump Up on doubles any more as it's normal access. Guard is still very high up the list. Your assertion that MB/PO would be the next two doubles, with precisely no information to back that up, belies your own prejudices more than anything. its not though is it, what other doubles are ever taken, the doubles would either be kill stack or ignore the double now, and doubles should never be ignored, also Jump up isnt a double any more but does it ever get taken on this player? the answer is no in my expereience.

- Amazon Thrower - again you make a bunch of assumptions. I do but it is based on a lot of experience, playing leagues and in MM envirnoments, I just don't see passing players developed anymore and when i do it is usually at the hands of a poor coach anyway

- Beastman - FA was nerfed, as you say, making it less attractive. Claw is still right up there, as is RSC (which, as a combo, is EXTREMELY powerful). I think your top 5 misses tackle out, but the point is still that they require a LOT of skills to be as good as they can be. well yeah top 5 tackle or frenzy, they will be interchanged, and while it does mix 5 skills that is not the problem the problem is it is the only combo you see people taking now on these teams. I cant remeber the last time I face Stand Firm tentacles block, or Shadowing prehensile tail and so on... actually I can, it was lrb4. but you get the point

- Bull Centaur - Old SF was very nice as a S skill, and dodge does part of what old SF did (help reduce turnovers) and can be used ot the same effect with BT, giving the BC back his all-important mobility. yes, but the point is there were other options, not just Dodge, as said even catch was taken a reasonable amout, again I havent seen Catch BUll Centaurs since LRB4

- Thrower - NoS is a single now. yes it is a single but the point is, people dont take it now, even though it used to be a double it was still more common than it is now

- HE catchers now make really good blitzers on a team which has only two. exaclty, detracting from a teams uniqueness, now making all the elf teams and elf positionals far more similar

- GR - VLL was changed dramatically iirc. FA was nerfed. exactly, detracting from variation, making skills less interesting and more vanilla

My point is that it may be the changes to the skills themselves which is a part of the issue. Some of the numbers you gave (runner with 58% dodge, Zon thrower with 76% SA) yup but the point is throwers were used to throw in the vast majority of cases indicate that there was a lack of variation even in LRB3. yeah i agree there still wasn't huge variation but it was more interesting and players got built for more specifric roles then, throwers got built to be out and out throwers, now players are all much of the same, except linemen who have now become pretty redundant on many of the teams

One of the things you don't mention is where there are similarities:
CW most common normal skills are Block, Guard, MB, Tackle, PO with Claw and RSC as a double. Now that is a killer. it's not as powerful as you think, MB didnt work with Claw RSC, also Pilling On sucked back then, it was Av re-roll only, the Claw RSC part was very nasty though, but that was soooo rare because of ageing, it was more common to see a player with one or the other and other players had mb block guard etc.., again making teams more intersting and players more unique. It was very very rare to see both RSC and Claw together and if ageing didn't knack them up then there was a natural counter as well - fouling, something CPOMB doesnt have

Dwarf Blitzer common normal skills are Guard, MB, Tackle
Human Blitzers - Guard, tackle, MB with dodge on a double.

Sure, some things are different, but some are the same. My point is that until we can have some info on how skills are distributed in CRP then we have no comparison and anything said about what we think the CRP numbers might be is pure speculation based on our own biased perceptions.

Some are the same and as said this will never change and some of it probably shouldn, stuff like dwarves spamming guard MB etc... this fits with their character, it makes it feel like you are playing with dwarves, too many of the other races have now become interchangable. I don't know if the skill selection survey could be done again. But if you ask Hitnogashi or Koadah they might be able to throw something together from BB and Ranked so it is the same sample as the one above. If you ask them to get give the top 5 skills taken and there percetages on singles and doubles for every positional they might be able to get it?

[dode74]

Just because you can't see an issue, doesn't mean it's not there.

I never said otherwise. What I'm trying to do is establish what the issue actually is. "I don't like it" isn't necessarily an issue with the game - it may be an issue with an individual's preferences and expectations with respect to the game instead. garion has gone some way towards providing tangible data.

@ garion
Most of what both of us wrote is argument about skills, and we do seem to agree in the end that data from CRP is needed. I'll see what I can get from Cyanide (since you won't see PB, for example, in CRP FUMBBL) as well as Koadah.

A couple of issues with what you said:

what other doubles are ever taken, the doubles would either be kill stack or ignore the double now, and doubles should never be ignored,

Why should doubles "never be ignored"? That's an assumption on your part.

the answer is no in my expereience.

but it is based on a lot of experience

Sorry, but your experience is not data. You need to take your assumptions out of this and look at the data itself.

I just don't see passing players developed anymore and when i do it is usually at the hands of a poor coach anyway

That's surely a passing issue rather than anything else. Passing is a hugely risky play in terms of the turnover risks.

the problem is it is the only combo you see people taking now on these teams. I cant remeber the last time I face Stand Firm tentacles block, or Shadowing prehensile tail and so on

In what environment? See, this is why we need to separate out MM and leagues. In MM, which is bash-heavy on all formats, outbashing the opponent will do you well as you get a numbers advantage. While elves can cope with such things other teams can't, so in an MM league where you are unlikely to face elves then going for bash is advantageous.

it's not as powerful as you think

I'm aware of the interactions, but plasmoid's numbers showed that LRB3 was FAR more bloody than CRP from these skills.

All that said, data is required before any comparison between LRB 3 and CRP skillchoices can be made. Much as I recognise you play a lot of games, your perception is not the same as data.

[garion]

again with the red, Im lazy

@ garion
Most of what both of us wrote is argument about skills, and we do seem to agree in the end that data from CRP is needed. I'll see what I can get from Cyanide (since you won't see PB, for example, in CRP FUMBBL) as well as Koadah.

A couple of issues with what you said:

what other doubles are ever taken, the doubles would either be kill stack or ignore the double now, and doubles should never be ignored,

Why should doubles "never be ignored"? That's an assumption on your part.
They shouldn't be ignored because they should be something to rejoice over, a special event, you should be rewarded for you good luck. This is what made the skilling up event so exciting, now it just isn't you just plod along.

I just don't see passing players developed anymore and when i do it is usually at the hands of a poor coach anyway

That's surely a passing issue rather than anything else. Passing is a hugely risky play in terms of the turnover risks. while that is true passing playes were still developed more often in lrb4 whether passing was done anymore is probably a no, but players certianly seemed (again I know data needed) to take passing skills more often than they do now, even when they required doubles and now don't

the problem is it is the only combo you see people taking now on these teams. I cant remeber the last time I face Stand Firm tentacles block, or Shadowing prehensile tail and so on

In what environment? See, this is why we need to separate out MM and leagues. In MM, which is bash-heavy on all formats, outbashing the opponent will do you well as you get a numbers advantage. While elves can cope with such things other teams can't, so in an MM league where you are unlikely to face elves then going for bash is advantageous. in both league and MM, I have recently quit all the leagues on fumbbl because even after two season things are already strating to shape up in the same way they have in MM and I have lost interest in the game as a result. In RL its harder to say, leagues i have played in are not long enough to see kill stack develop properly etc..., also unlikley to face elves in MM ? Im not sure I mostly played elves, because apart from the CPOMB teams they have done the best out of this rule set.

it's not as powerful as you think

I'm aware of the interactions, but plasmoid's numbers showed that LRB3 was FAR more bloody than CRP from these skills. In theory bowl; possibly, and maybe LRB4 was more bloody all round, personally I suspect it is about the same, the nerf to fouling and cas caused by that has probably been replaced by the new kill stack. But I don't care about cas, that doesnt bother me, I am happy for people to die, the problem is it all comes from one source now, while it used to come from all angles.


All that said, data is required before any comparison between LRB 3 and CRP skillchoices can be made. Much as I recognise you play a lot of games, your perception is not the same as data.

No its not your right, but I do value the opinion of people like darkson, Joemaji and so on and if many experienced coaches think something is wrong with team development, the liklihood is there is. But yes, get the data from MM and set it out the same as the link I provided. It should be interesting.

[dode74]

Yes, you are lazy

- Doubles are still a special event, but that doesn't mean that they should always be taken. I set out a plan for the type of players I would like in my team, and if a double or stat comes up and it doesn't fit with the plan or open up better options then I don't take it. An example would be that I recently turned down +AG on a lineelf. That would be because I already have a witch and a blitzer with AG5 (and leap), and I am lacking in wrodge lineelves (only have one other and he is MA5). For the team, dodge was a better skill selection imo.

- Without further data we can't really say anything about the passing skills.

- Data shows that elves are played less than many other races despite having some of the best win%. I agree that elves have done really well out of the rules changes, but as I said before, people don't play MM to win, they play to survive or to team build. In OCC all the CPOMB teams have started to slide down the rankings with the new races coming in - skilling CPOMB for the original cyanide 9 worked, but is failing against the newer teams due to an average lower AV. Again, without data we're both talking perception.

- CPOMB is theorybowl as well. I agree fouling needs a boost though, but that is off topic.

- I value people's opinions as well, and I am not dismissing them. I am saying that opinion is often coloured by expectation, and data can show surprising variance from that expectation. As for "wrong with team development", wrong is not the same as different. Team development may be different, but that doesn't make it wrong.

[garion]

When you get the skill data. display top 5 for each player. The skills aren oredered as 1st pick 2nd pick etc.. it is just percentage that the skill has been taken on a player we are looking for

[Chris]

Leader is on throwers because in American football they are the captains I believe and make the plays. Makes sense to me that a player trained to look for openings to throw the ball would have better awareness of the game and grant a re-roll. Certainly better than the lineman it used to go on.

I agree with costing skills differently, though I think also they should have a couple of tweaks to their effect. I never liked traits much.

Two things though, 1 I don't agree with the list above and 2 this would have a big effect on some teams. The former can simple be argued over, the latter would impact on the game in unforseen ways (dwarves would I suspect become worse at high tv levels and so on).

The skill 'nerfs' were I think in general good. They benifited some teams far more than others - e.g. didn't very long legs mean increased move? Hence why all the one turning gutter runners loved it.

Of course fix this and you would still have the massive effect ag4 has over ag 3 at very high tvs, or block focused killer teams do at the same high tv levels.

Of the list above you have

10k list
Pass
Accurate - how is this not popular?
Fend - again as a late skill pick quite popular
Sneaky Git - well a bit naff, would give a boost
Very Long Legs - currently a slann skill or the preserve of players who can get leap and mutations. Making it cheaper just seems to buff Skaven
Kick Off Return - A useful skill, especially in late development but even early development with some teams
Pass Block - likewise
Sprint - Some teams like it, certainly on par with many of the 20k skills below
Diving Catch - How is this not good for certain players on most teams?
Sure Feet - Again numerous builds like it as a mid development skill
Dump off - Excellent for ag 4 teams, not so good for anyone else. Why buff them?
Safe Throw - For thrower they will take it eventually if they are interested in throwing the ball. Great with dump off and by extension even ag 4.

20k list
Block, Dodge - does this not change relative value as teams develope? Dodge is great on starting teams, less use on very developed leagues as tackle piles up.
Wrestle,
Tackle,
MB,
Kick - You see this less than lots of the skills in your 10k bracket, yet all teams like to have one...
Dirty Player,
Catch - shirley of less value than diving catch given most players that want such skills start with it? Not sure if you see pass taken as much as this as well.
Tentacles
Prehensile Tail
Two Heads
Extra Arms
Big Hand - worth more than two heads/extra arms surely? The big handed gutter runner is a staple...
Side Step - regarded by most as better than stand firm
Nerves of Steel
Shadowing
Stand Firm - far more common than it otherwise might be because of the good synergy with strength teams
Multi Block
Horns - I would rate this better than dauntless, however I recognise that only gutter runners and goblins really like to take it (well and the odd marauder)
Dauntless
Sure Hands
Strong Arm
Jump Up
Leader

30k list
+Ma - +MA is worth more than +av
+Av
Claw - well yes, is a rather good option!
Diving Tackle - I would have thought tackle is better than diving tackle, more common too
Leap - Yes its good, but few players can use it well
Guard - Yes, though would impact on teams that rely on it (dwarves, humans and then latter on bash in general)
Pilling On - Could do with a nerf



I think you would combine any costing of skills with a few tweaks to them. I would look at

Sneaky Git - Some sort of boost in addition to the ability to ignore doubles - allows the player to give +1 to a foul even if marked when assisting/allows the player to commit a foul with +1 to the armour roll/only ejected if target removed from the pitch and rolled a double

Pilling On - Decide wether or not to use piling on after the armour roll is made. No modifiers apply to armour or injury on the second roll, player placed prone. - Skill shouldn't stack, its not a finese move you and simply trying to squash someone!

Disturbing presence/foul appearance - could combine them for a good skill? (Otherwise have them in the cheap bracket)


Another idea to consider is if each skill should cost an additional increment. So skill 1 20k, skill two 30k, skill 3 40k, 90k total instead of 60k. Would impact teams that skill and rely on specific players over teams that develope evenly (elves).

[tchatter]

A change like this would cause a "snowball effect" throughout a lot of the other parts of the game; inducements for example would need re-balancing as would many other things. As someone replied earlier... up to 5 skills for one improvement roll!!?? Really....???

I can see where the "problem" is, its that if you go and say search the internet for "Blood Bowl Chaos Tactics playbook" you get a few cookie-cutter ways of building your Chaos team. Anyone who is in an MM environment that doesn't use a cookie-cutter build will typically not fair as well. TV efficiency and all.

So rather than give players a TON of extra skills on the cheap, why not take a look at the skills that are deemed unworthy and try to give them a little buff?? That wouldn't cause the rest of the rules heartburn and require any tweaking in other parts of the game.

[mattgslater]

I think there are now two conversations going on.

1) The core problem, being that there is a dominant LRB6 playstyle (worse, not a very fun one), and the credible alternatives mostly revolve around power skills, yielding just a few types of teams that can really win. This is a real concern in my book, and is a major flaw with the new ruleset.

2) Joe's idea for solving the problem. I think the consensus here has been that it needs toning down, but a rule in the same spirit might solve the problem or help solve the problem.

Observations:
* Skills stacking, and generally speaking intelligent skill selection, will yield a more than arithmetic impact.
* Some stacks are better than others: namely, killer skill stacks, which go too high with too much depth of quality (one too many good skills) if you have GSM access, and power skill stacks, which rely on pure skill quality and usually take doubles to get any real depth.
* Joe's plan is, at its root, to encourage stacking skills that don't fit into killer or power skill stacks.
* Since the gain for multiple stacking skills is to some degree exponential, even just one extra skill can carry great value, especially after the attendant power skills have been taken.

Me, I think a 2-part solution is appropriate: one, a way to make low-priority skills more accessible, and the other, a way to account for the power of skill stacks in TV calculation.

A) Make a way for players to get bulk rate on "C"-grade skills, but not to the tune of more than one free C skill on a player with at least one more; a one-time two-for-one would be an easy way to do it (take one skill from this list, get one skill free). In that case, you could have two lists: one for useful-but-not-optimal skills you want to encourage, and one for cheesy skills that aren't worth much. So you couldn't get Pro/Fend or Sure Feet/Sprint for one skill, but you could get Fend/Pass Block or Sure Feet/Sneaky Git. This also reduces abuse by, say, Vampires (who just looooove Pro, and have so many different access categories), or big guys on second doubles, who are cool enough with Pro, let alone free Grab or Sure Feet.

B) Add value to improvements beyond the second, say 20k for the first two, and 30k each after that?

Here's how I'd break down the list in A, using the original list as a template.

If you take: Fend, Kickoff Return, Pass Block, Pro, Shadowing, Catch, Sprint, Sure Feet, Diving Catch, Hail Mary Pass, Nerves of Steel, Dump Off, Grab, Thick Skull, Strong Arm, Extra Arms, Disturbing Presence, Prehensile Tail, Very Long Legs, Horns, or Big Hand

you may also take: Pass Block, Shadowing, Diving Catch, Sprint, Hail Mary Pass, Dump Off, Thick Skull, Strong Arm, Disturbing Presence, Extra Arms, Prehensile Tail, Horns, or Very Long Legs.

once in your career, free of TV add, on the next improvement. You must be allowed to take your selection on the new roll (no doubles skills on the second list unless the next improvement is doubles).

Sprint/SF is possible, but you need SF first, then you have to wait to go up again before you can get free Sprint. Same with Big Hand and Extra Arms: you need part B first, but if you take it, you can get part A free on the next roll. (If you don't like that, put Big Hand on the second list.) If you double with a big guy and take Pro, your next (free) skill can be Thick Skull or Strong Arm, not Grab or any doubles-only skill (unless you double again, and then you just deserve your awesomeness).

[dode74]

You see, that's my point - we hadn't actually identified the problem that Joe was trying to solve. The problem you have defined is actually in two parts: first that there is a dominant playstyle (and the stats seem to indicate otherwise), and second that the game is less fun (which is a matter of taste). Joe appears to be saying that there is less variation in skills taken and wanted to improve that.

Regarding your observations:
- Skill stacking gives an arithmetic bonus where the skills are additive (e.g. Strong arm and accurate), but a multiplicative bonus where the skills add and allow rerolls (pass plus strong arm and accurate).
- Agree that some stacks are better, possibly due to the nature of some of the additive and reroll skills (MB and PO can be used on either roll, for example). "Better" does not necessarily mean "too good" though.
- To me, Joe's plan appeared to be to increase the incentive to take allegedly rarely taken skills. Your plan, however, (if I understand it correctly) may herald a sensible approach.
- As per my first point above, it is the reroll skills which cause multiplicative increases in capability. In particular, skills which add and allow rerolls cause a multiplicative improvement.

[neverworking]

Joe appears to be saying that there is less variation in skills taken and wanted to improve that.

This statement seems to come up a few times in your posts. I would say its not terribly critical to state there is less variation than some other rule set. If there is less variation than another rule set you can conclude recent changes in the rules have lead to a more cookie cutter approach in development, however its at least my point of view that variation is a desired outcome to begin with because it will lead to more varied and interesting game play. If skill selection frequency avoids a collection of skills that are intended to be good skills then something should be done to encourage the use of skills that aren't as appealing as they are intended to be.

Joe has proposed encouraging it through a change in skill values which I think makes intuitive sense. While as some have pointed out that every roster would need to be looked at, it should be noted that the old "cost formula" for starting builds actually counted some skills as 30k skills and others as 10k skills. Over time the accepted list of 30k and 10k skills has changed slightly, but it seems very logical to assume that if the original formula for designing a player put different costs on skills that future skill selection should encompass it as well. It is not coincidental that some of the teams that are rather successful at low TV start out with best early access to 30k skills, but begin to falter later in development since they are effectively selecting the 20k skills at stages when other teams are selecting the 30k skills.

It would seem to me the goal would be to develop any set of changes that brought the frequency a skill appears on a developed (and successful?) team to a certain level or alternatively the frequency it appears on a player at some experience level (e.g. found on a team with TV1500-1800; taken on a player in first 3 skills) . I would probably approach it from a probability a team has a skill on it rather than total count of skills since some skills are universally useful on all players whereas others provide diminishing returns as the quantities rise (kick, kick off return, pass, etc.), but it may be wise to look at the metric from a total frequency in first 3 or 4 skills for each player type as well. While we all have personal bias on what should be a 10k or 30k skill, the actual frequency of the skills should tell us all something is over or under costed more than anything, and a query of the FUMBBL or Cyanide database would be the best starting point for the list. Some allowance would need to be made for the fact that some teams have little access to a skill that many would want or that some skills are only applicable to some teams, but on the whole, getting each skill's selection frequency within a targeted range would be the empirical goal Dode is probably seeking as the current environment is measurable and future success could also be measured.

[mattgslater]

As per my first point above, it is the reroll skills which cause multiplicative increases in capability. In particular, skills which add and allow rerolls cause a multiplicative improvement.

See, I think you can get more than additive improvement by combining additive skills. Having a guy with Claw + MB is better than having a guy with Claw and a guy with MB, because you can put both skills in the same place at the same time. Similarly, given a pair of +1 modifiers on a 4+ roll, it's the second one that really makes the roll safe to attempt. This is where spatial relationships and skill distribution really come into play, so that skills whose effects may seem additive are actually exponential.

On top of that, there's exponential value when considering the multiple aspects of a die roll. If Block adds a third to your knockdown chance and MB doubles your Cas chance, then Block/MB adds 5/3 to your Cas chance (2x(4/3)-1).

[dode74]

This statement seems to come up a few times in your posts. I would say its not terribly critical to state there is less variation than some other rule set.

I haven't said that there isn't. What I have said is that we don't yet have the data to say it is the case. If it does prove to be the case then there are several methods along which we can advance, including changing costs and buffing or nerfing certain skills.

Your last paragraph seems sound as a method to me. Other methods include a straight comparison with other rulesets, or perhaps some sort of decision on what "not taken enough" actually is. I'm not certain how useful it will be to look at things by race since races have different playstyles, so I'd be tempted to look at the data in the round initially (which is what you are saying).

getting each skill's selection frequency within a targeted range would be the empirical goal Dode is probably seeking as the current environment is measurable and future success could also be measured.

Possibly. Deciding on what that targeted range should be will be difficult though. I suspect that a better metric might be to make the decisions over which skill to take next a tough one, or to have a wider range of "tertiary" (i.e. 3rd pick) skills which are more closely chosen by frequency on various races.

See, I think you can get more than additive improvement by combining additive skills. Having a guy with Claw + MB is better than having a guy with Claw and a guy with MB, because you can put both skills in the same place at the same time. Similarly, given a pair of +1 modifiers on a 4+ roll, it's the second one that really makes the roll safe to attempt.

Mathematically they are additive. Looking at your 4+ roll, each +1 is still a +1 - the odds of success or failure change by the same amount as a proportion of the whole.

On top of that, there's exponential value when considering the multiple aspects of a die roll. If Block adds a third to your knockdown chance and MB doubles your Cas chance, then Block/MB adds 5/3 to your Cas chance (2x(4/3)-1).

I see where you're coming from, but MB never doubles your cas chance (10+ to 9+ is from 6/36 to 10/36), it adds 2/3 to it. Looking at skill interaction across multiple dice rolls (block then armour then injury) will always be multiplicative though, so your point is fair but I think if you start looking that deeply into it you're actually looking at a whole game (e.g. sure feet gives a GFI RR which puts you in place for a short instead of long pass with a pass RR which allows a catcher to make a catch RR etc etc)!