PinkHorror's Team Model

[GalakStarscraper]

Okay ... I've been working with PinkHorror to take what I've seen about the MBBL2 and make some revisions to the league to balance out the few remaining bumps (like the Albion roster) for Season 3.

To this effect, PinkHorror made a very cool team analysis MODEL (not formula). Its looks at all the aspects of a team ... stats, skills, access, cost, rerolls, etc. to review overall team balance. Basically the rule was simple. The Power of the team and the Value of the team should NOT exceed 80 for either calculated value. Perferably only ONE of the values of Power or Value should exceed 70, but we decided that rather than disallow such teams that they fall in the category more of what carefully for possible future revision change needs.

So here are the goods ... PinkHorror's team analysis model can be downloaded here ... it has all the player from both GW Official and the MBBL2 allowed teams in it along with all the extra skills from the MBBL2.

http://www.midgardbb.com/JervisRules/teamanalysis.zip

The 76 team rosters allowed in the MBBL2 can be found at:
http://www.midgardbb.com/BloodBowlTeams.html

The MBBL2 has been an interesting experiment. Many thought that with 76 rosters that the internet based concepts would overrun the league ... however with the 2nd season drawing to a close ... again the GW rosters are doing better than the internet concepts. The lone exception the Albion team is being completely re-written for Season 3.

Anyway for anyone interested here are the stats on the MBBL2 rosters based on PinkHorror's model:
PH can modify my descriptions ...
Power is the overall rating of the ability of the team
Value is an analysis of the price you pay for the power you get
Blocking, Throwing, and Running are all measurements of the ability of the team in these key areas.
Potential is the growth potential of the team
Pricing is a modification of my old discount formula ... how much is the team under/over costed for the player stats and starting skills.

Code: Select all

            NAME          Power  Value Blocking Throwing Running Potential Pricing
Albion Wanderers          71.68  72.81   80.87   53.59    61.25    66.83     52.07
Amazon                    66.38  60.22   66.99   54.85    66.34    69.86     57.04
Araby - Holy              65.81  59.06   72.42   53.37    53.96    66.66     52.46
Araby - Unholy            71.86  57.48   83.97   53.30    52.59    70.85     48.85
Barbarian                 57.94  22.10   96.13   18.68    31.79    68.46     28.60
Border Princes            68.42  52.62   70.19   59.66    63.97    68.31     52.23
Bretonnian                73.22  60.58   76.05   59.61    70.58    70.00     51.38
Bright Crusaders          66.82  47.40   67.25   59.25    65.41    67.23     50.99
Chaos                     74.22  46.56   86.07   50.19    51.64    84.09     45.87
Chaos Beastman            70.34  59.70   78.55   50.21    53.26    80.91     48.35
Chaos Dwarf               75.06  76.54   97.33   43.06    49.13    75.55     54.00
Chaos Halfling - Immortal 43.03  39.23   42.73   37.31    39.25    55.57     39.89
Chaos Halfling - Strength 49.27  47.10   58.00   32.66    39.03    57.51     41.82
Chaos Warrior             79.09  48.99   91.99   51.42    56.49    85.62     49.94
Dark Elf                  79.88  46.89   74.64   74.45    77.67    81.50     52.10
Draconian                 74.94  49.53   86.98   53.09    60.86    68.70     44.62
Dwarf                     76.52  63.02  100.45   43.95    47.02    77.74     56.37
Elementalist              75.94  60.17   86.76   51.58    59.84    77.64     54.29
Evil Gits                 78.08  58.84   86.45   58.69    61.39    77.77     50.56
Fimir                     78.66  47.23  108.11   35.83    47.76    87.03     46.53
Forest Folk               74.65  60.21   84.20   53.31    65.30    68.90     51.17
Forest Goblin             77.60  59.47   83.55   49.88    75.76    79.52     53.29
Gnome                     62.94  77.67   70.53   42.32    63.75    60.50     58.21
Goblin                    51.49  75.71   50.66   37.51    60.61    52.20     55.61
Golem                     77.58  52.91   95.69   49.19    57.73    70.35     46.99
Halfling                  46.54  57.88   48.67   33.48    46.73    56.34     54.25
Half-Orc                  76.08  51.63  101.99   39.99    50.47    73.83     50.35
Halloweenies              79.41  73.11   98.66   49.81    52.18    79.25     51.75
Heroes of Law             79.06  57.29   87.12   63.56    61.90    74.26     52.71
High Elf                  74.59  44.99   59.64   74.02    82.06    80.42     52.78
Hobgoblin                 63.16  45.53   67.17   49.43    60.98    63.85     46.86
Human                     66.82  47.40   67.25   59.25    65.41    67.23     50.99
Khemri                    73.91  76.98  105.62   39.25    38.36    76.63     51.41
Khorne                    79.64  41.71   93.11   50.07    56.42    88.23     43.53
Kislev                    76.27  69.92   85.67   54.37    62.41    75.05     55.10
Lizardman                 72.15  50.99   87.57   38.67    62.84    73.57     50.20
Lycanthrope               66.04  43.88   84.13   29.72    54.82    76.58     49.00
Minotaur                  65.89  25.64  118.02   17.75    27.28    92.19     32.11
Mongrel Horde               ???    ???     ???     ???      ???      ???       ???
Night Goblin              54.16  76.16   60.58   37.51    52.02    56.67     51.63
Nipponese                 76.28  53.07   81.48   57.21    68.35    75.74     49.53
Norse                     78.01  65.13   92.75   53.52    60.83    69.37     55.68
Nurgle                    79.91  42.13   97.10   49.99    49.55    90.10     43.83
Nurgle's Rotters          75.58  49.44   86.94   51.15    55.00    82.60     47.43
Ogre                      70.65  24.16  120.73   21.95    33.04    84.14     30.78
Orc                       75.23  67.02   88.12   54.57    55.24    72.25     52.39
Pixie                     57.65  78.75   69.50   42.71    39.72    61.58     53.34
Robot                       ???    ???     ???     ???      ???      ???       ???
Savage Orc                79.57  59.64   95.98   50.41    58.25    76.95     50.76
Shallya                   68.90  64.02   62.93   66.22    72.15    68.57     55.99
Sisters of Sigmar         64.96  46.48   70.24   56.68    56.60    61.12     49.64
Skaven - Clan Axme        62.56  61.74   60.45   58.11    60.19    71.73     53.00
Skaven-Eshin              61.66  57.13   41.95   66.01    75.05    78.28     54.64
Skaven-Mors               61.15  53.32   55.34   58.91    62.47    69.57     52.46
Skaven-Moulder            67.60  68.98   68.81   56.96    57.36    79.26     53.25
Skaven-Pestilens          53.41  61.86   44.63   56.99    57.45    60.64     53.81
Skaven-Skrye              53.41  61.86   44.63   56.99    57.45    60.64     53.81
Skaven-United             69.23  54.56   61.96   64.23    72.06    75.04     52.87
Slaanesh                  78.90  43.20   89.01   55.57    56.27    86.66     45.52
Slaanesh Elf              72.00  75.90   67.41   66.70    70.32    79.19     56.59
Slann                     74.01  48.01   71.16   61.75    77.37    73.17     51.48
Snotling                  47.67  79.78   65.06   25.72    28.69    57.72     51.95
Spirit                    66.77  60.04   68.40   58.68    65.45    64.18     52.59
Treeman                   73.55  24.86   95.69   42.57    49.36    72.01     27.88
Tzeentch                  79.25  43.35   88.50   56.28    58.01    86.48     45.47
Undead                    73.59  59.01   85.80   52.69    57.45    69.52     50.08
Valkyrie                  78.11  75.72   87.76   57.19    69.04    67.98     57.40
Vampire-Blood Dragon      78.29  71.58   96.89   52.03    52.36    74.07     52.42
Vampire-Lahmian           79.98  76.43   85.79   71.82    65.39    69.25     51.12
Vampire-Necrarch          78.67  76.49   95.61   59.59    60.53    59.71     55.85
Vampire-Strigoi           79.81  66.82   96.29   52.34    61.37    70.79     51.53
Vampire-Von Carstein      77.35  75.94   84.95   63.03    65.61    67.95     51.71
Wee Folk                  48.57  44.25   49.28   39.48    49.63    55.59     44.41
Wild Cards                  ???    ???     ???     ???      ???      ???       ???
Witches of Selene         69.20  53.24   71.69   54.38    66.84    71.60     52.81
Wood Elf                  79.91  52.34   61.39   74.94    92.87    83.91     54.97

[Dangerous Dave]

Galak and PH

This is all very impressive!

However, as I'm sure you will acknowledge, it is very subjective and depends upon what value is assigned to each skill etc.

I have taken a quick look at the spreadsheet and I feel that there is something missing (it may be that its there but the spreadsheet is time consuming to follow to understand where all the value is created):-

Is there anywhere which assigns a value to how hard the team is to block - a sort of a counter value to the Blocking value?

The reason I ask is that on comparing 2 similar positions, the spreadsheet gives a different answer to that expected (by me):-


Name Cost Discount

Amazon Linewoman 50,000 6,836
Norse Lineman 50,000 9,741


The Norse Lineman is rated more highly than the Amazon linewoman primarily because the Norse lino starts with Block. Now after 1 skill roll, the Amazon Lino can take Block and the Norse lino would need to roll a double to keep pace. That to me suggests that the Amazon lino should be priced higher than the Norse lino.


In addition, a lot of the value is very subjective. For example the Chill ability is rated at 1.3 (for blocking). This gives a discount of 8,374 on the price of 90k for a Wraith. Now if I change the rating to 1.35 this provides a premium of 5,168. Now I don't know how the 1.3 is used in the formula and I also have no idea what sort of value should be given to Chill. However, for comparison the following ratings are given for blocking:-


Mighty Blow 1.14
Piling On 1.18
Guard 1.95
Block 1.35


Using a Chill rating of 1.3 the Wraith has a Block rating of:-

Wraith 83.36


Using 1.25, the Wraith has a Block rating of:-

Wraith 80.16


Using 1.35, the Wraith has a Block rating of:-

Wraith 86.57



Maybe this is small but my question is how much is Chill worth? Is it worth more than Mighty Blow? Is it only marginally less valuable than Block? I am not saying I know the answer - that's the point, no one does.

The point of this post is that I would caution about relying on any of the output. Sure its a very good effort and a lot of time has been spent on it - so kudos to PH for that. But at the end of the day the output will always depend upon whatever the inputs are... and these will always be subjective.


So this is a useful tool... but.... please remember that if the output says one team is rated at 78 and another is 75 that doesn't mean the one rated at 78 is better. It simply means using the subjective decisions made that it is better - a very important distinction.


Dave

[Dangerous Dave]

Oh sorry one other point. I can't see anywhere where Secret Weapon rolls are taken into account. For example the Storm Giant is priced at 140k and a discount of 2392. So all well and good. However, since he is a Secret Weapon at 7+ (ie will be sent off 58% of the time after each drive) - the odds are that he will play less than one half. OFAB is valued at 0.77 across the board (Blocking, Throwing and Running) - so I would say SW 7+ should be 0.7 or maybe 0.75.

If I add the OFAB trait to the Storm Giant, the Blocking value goes:-


From 198

To 152


and he has a premium of 24,609. So an approximate change of 30k (24.6 +2.4 plus a bit since SW 7+ is worse than OFAB) for the SW 7+.


Ok this is a one off and rare player.. but it just goes to show how much a value can change.


Dave

[GalakStarscraper]

PH will have his own response I'm sure ... here's my quick ones.

1) The Storm Giant has the Unique trait in the rater ... I decided that the negative value of Unique was equal to a Penalty roll of 7+ so it was accounted for in the calculations.

2) PH and I will both readily agree that the spreadsheet is subjective. Anyone using any one thing as a be all end all method is wrong. I learned this the 1st season of the MBBL2 when I tried to force the 50-60k team discount formula on all teams. That was a bad idea. HOWEVER, I will be honest with you every team that I entered into PH's spreadsheet that ended up with more then 80k in either Power or Value I fully agreed was too powerful ... ie I didn't have a single time where the a team exceeded 80k in the stats that I didn't feel that the team need revision.

3) I fully agree that Power IS NOT an end all indicator of one team being better than another. Each stat is very important to me. Look at the Power rating of the Wee Folk ... I came really really close to beating a Wood Elf team 50 TR points higher than my Wee Folk team last round and the Wood Elf team has a power rating that is a feather shy of over 80. Another stat that I didn't post that I look at is (Blocking+Throwing+Running)/3 which gives a different look compared to Power ... sorting on this will give you a much different look at a team than a sort by Power. Bottom line on this one: Power is not an end-all number to use to compare teams ... however PH and I both agree that if your Power rating is over 80 ... your team is TOO good.

4) The values on the abilities are definitely subjective ... In fact if you look at PH's notes on the front page ... he is very up front that the sheet is his based on his ideas and is thus highly subjective. There is NO WAY to create a model for BB teams that would not be subjective.

5) The sheet was a final check on my changes for Season 3. It should be noted that PH's sheet is ALMOST all case confirmed the teams that I had ALREADY identified as needing modified. A few minor tweaks resulted from messing with PH's sheet. I guess what I am trying to stress is that I found PH's sheet a very valuable tool for making sure that a team wasn't TOO good. In Blood Bowl, very rarely is a new team suggested that is too weak ... its generally an ill-conceived juggernaut. PH's worksheet does a really good job of pointing out such teams which is 90% of the battle with new BB teams.

I think that's all my thoughts. Anyway, PH has told me he still has some stuff to add. If you think the scores on any one ability is out of whack lets talk about it ... but I REALLY like PH's attempt to give each skill a different value based on the type of player possessing the skill.

Galak

[Pink Horror]

Oh, I'm glad to see this is online. I was working with the daemon teams when the electricity went out and so I promptly bought a UPS, but I haven't gotten back to the spreadsheet because of that frustration. Looking at this list Galak has, I guess he got to them already. I'll have to see what he did.

Now, Dangerous Dave, I'll try to answer your questions.

Blocking and counter-blocking were combined into the same thing. Being able to block was considered better than being able to only absorb blocks. I would like to add a fourth category when I rework the whole spreadsheet and I'd call it Defense. Then there'd be a better place for bonuses such as the ones from shadowing, kick, tentacles, and very long legs. I'm also planning on having not just constants for each skill, but also room for instructions on special combinations so I don't have to hard-code those.

From earlier experience with GW's pricing scheme, I didn't include future development as part of a player's price. So, maybe the Amazon gets better later, but she starts out worse. The difference is in the Growth (a player's version of Potential). An Amazon Linewoman has a Growth above 57, and the Norse Lineman's is under 47. These numbers become part of the Power of a team.

I got most of the values by beginning with the assumption that most players had fair prices and I worked backwards. Chill seems to be worth a lot when I looked at it that way. It is, after all, a free action that can steal a turn from one opposing player. Mighty Blow, on the other hand, gets most of its worth from your other blocking abiity. The players with mighty blow are all great blockers without mighty blow, so it's a smaller multiplier. I tested different numbers with Block for a long time on several different players which I thought had reasonable prices. That's how this usually went - I take a player which I feel is priced fairly and then I mess with the constants until that player works out to the right numbers.

Chill came out to 1.3 for a few reasons, some fair and some not so fair. First, I didn't want to spend a lot of time on chill because only one team has it. That meant I tried 3 or 4 different numbers, picked my favorite, and moved on. The key reason I stuck with 1.3 is that the spectre was overpriced ~10k and the wraith was underpriced ~10k with that number. Looking at their stats, that made sense, so I was happy. 10k here or there doesn't matter much. The system is flexible.

Secret Weapon rolls are only on a few players that all have some trait or another that I reduced in accordance with that defect. I was stingy on purpose with most negative characteristics, so the Storm Giant would probably have gotten between .75 and 0.8 with me. The first drive is the most important, after all, and you might stretch it out to keep him on the field.

It looks like Galak used a proper substitute to equal that penalty roll so I'm okay with whatever he got.

The big goal of my sheet was to make teams that are really focused on, say, beating opponents into the dirt would stand out as the powerhouses they are instead of seeming balanced because everyone has an accurate price. If you design a team that has a little of everything and no super players you should be okay.



Pink Horror

[Mestari]

The idea seems excellent, and like an excellent starting point for aiming towards the "ideal BB team evaluator engine". Which naturally will never be reached, but we can converge somewhere near it

As a future development ideas:

More emphasis on availability of skill categories, skill combos and development.

And along those lines, try to develop two sets of numbers for each team:
-in a tournament (from no to little development)
-in a league (long-term success of the team)

Great job =)

[Dangerous Dave]

Galak's points:-

1) The Storm Giant has the Unique trait in the rater ... I decided that the negative value of Unique was equal to a Penalty roll of 7+ so it was accounted for in the calculations.

OK! I wondered what that did!

2) PH and I will both readily agree that the spreadsheet is subjective. Anyone using any one thing as a be all end all method is wrong. I learned this the 1st season of the MBBL2 when I tried to force the 50-60k team discount formula on all teams. That was a bad idea. HOWEVER, I will be honest with you every team that I entered into PH's spreadsheet that ended up with more then 80k in either Power or Value I fully agreed was too powerful ... ie I didn't have a single time where the a team exceeded 80k in the stats that I didn't feel that the team need revision.

I agree using the spreadsheet is a good guide - but shouldn't be the only determinant of a team's power / value etc.


The rest!

I agree it is useful and it does point out uber teams....... basically my post was just to say - use with caution - it is one method of looking at teams... but is not the only one.


PH

Blocking and counter-blocking were combined into the same thing. Being able to block was considered better than being able to only absorb blocks. I would like to add a fourth category when I rework the whole spreadsheet and I'd call it Defense. Then there'd be a better place for bonuses such as the ones from shadowing, kick, tentacles, and very long legs. I'm also planning on having not just constants for each skill, but also room for instructions on special combinations so I don't have to hard-code those.

Good - I think that this will add to the results so far.

From earlier experience with GW's pricing scheme, I didn't include future development as part of a player's price. So, maybe the Amazon gets better later, but she starts out worse. The difference is in the Growth (a player's version of Potential). An Amazon Linewoman has a Growth above 57, and the Norse Lineman's is under 47. These numbers become part of the Power of a team.

OK I understand that. So future development is part of the Team's values but not the individual player's. As to whether Dodge is worse than Block.... I'm not sure. For a rookie team against another rookie team dodge is often better (unless the opponent has lots of Block or Tackle) - since when the opponent blocks he will have to roll a POW to get your player over. Of course we are into subjectivity again here.


Finally, I have entered a team in the Test part of the spreadsheet. I won't say what it is...... However, the values the spreadsheet produces are:-


Blocking 95.32
Throwing 56.28
Running 57.07
Potential 77.20
Pricing 52.58
Power 80.67
Cost 1,130,000
Value 75.16


I have my own views on this team - what would you say?


Dave

[Dangerous Dave]

One additional point for consideration. As far as I can tell, no allowance has been made for allies which can be added to the team.

For example, if I add an ally (eg a Norse Blitzer) to the Dwarf team the Power and Value ratings are:-

Power 77.79 (from 76.52)

Value 62.96 (from 63.02)

So no major problem. However, if I add an Amazon blitzer to the Wood Elf team, the ratings are:-

Power 81.41 (from 79.91)

Value 52.13 (from 52.34)

So this goes over the 80 threshold for Power. Of course allies can't use TRRs so the increase is less than this, however, in determining the overall Power (and all the other attributes too), allies should surely be considered. If all teams had the same ally access then it wouldn't be an issue - however, they don't so I think it is an issue.


Dave

[GalakStarscraper]

The team you presented Dave is too powerful ... I have a feeling that I know what it is and well ... if that's true I'm sorry to hear that that is the stats. I won't allow it in the MBBL2 in that state.

As for allies .... I only focused on the base teams. I agree that allies can boost a team like the Wood Elves over Power 80 which boils down to two things.

1) Most people are right when they say that allies are evil.
2) For Season 3 of the MBBL2, I'm just getting rid of the stadiums. Although I came really close to getting rid of allies as well. I didn't because a LOT of members asked me to keep them around. Out of the 80 teams in the MBBL2 only 2 that I can think of have more than 1 ally, and most teams don't have any allies. So I didn't include allies in my calculations for the MBBL2 because at this point I'm not sure they'll make it to Season 4.

Now of course if you want to rake me over the coals on using PH's model for the MBBL2 because allies cause teams to go over 80 Power .... I'll gladly remove allies from the the league for Season 3 and direct all the personal e-mails that I got the last time I mentioned I'd do this to you to respond to.

Galak

[Dragonman]

Sorry for this really random post.

Horay for Wild Cards (and robot and mongrel) teams there so strange they can't even be given any values. So beware they could be the strongest teams out there. If your lucky.


roll those sixes.

[Dangerous Dave]

On the Allies... No I don't want all the emails.... but if every team is reasonably balanced without them...... it seems to make sense to get rid.....


Dave

[Pink Horror]

I'll bet that plenty of teams can be designed that go over 80 power and still aren't really that strong. You can probably also design a team that actually gets under the bar and is still overpowered. This doesn't mean the system isn't useful. If you run a league like Galak's, you ought to have some way to evaluate new teams. This spreadsheet makes it easier. If your team is over 80, make some adjustments. There's plenty of room. That team you presented is one minor change away from fitting in. It would still get some scrutiny because of the high value, and the individual players should be checked for pricing, but that's not much of a problem.

In the MBBL2 every team has its intangibles that aren't calculated. There's spells, secret weapons, special team rules, bionics, and allies. I can't account for all these options yet. I hope that they're balanced with team powers. The wood elfs don't get many allies and they have no secret weapons, right? That sounds like below average for extras.

The Wild Cards are certainly able to produce a great team with some luck. I don't know how to measure that yet. Measuring the Robots is a little easier - just think up a combination of 16 players (no more than 6 different kinds, though) and enter them in. I haven't found any that are that strong yet. The Mongrel Horde seems to be a batch of real losers, so I'm not concerned with them.



Pink Horror

[GalakStarscraper]

Sorry for this really random post.

Horay for Wild Cards (and robot and mongrel) teams there so strange they can't even be given any values. So beware they could be the strongest teams out there. If your lucky.

roll those sixes.

Zombie once said that the Wild Cards team was one of the most broken team concepts that he has ever seen ... .... since that statement there have been 4 Wild Cards teams in the MBBL2's history. The total record for the 4 teams is 11-8-2 .... so far the team has not been a problem in terms of being broken at all. Its a wacked out team to play but the lack of Agility and Passing skills makes it a basher team with an inflated TR.

The Robot teams may need re-looked at .. in the current LRB environment ... cash flow may make the team too difficult to develop. Definitely not a team to really fear.

Finally the Mongrel Horde... so far 3 Mongrel Horde teams have played in the MBBL2... combined record 1-8 .... again not a team that I lose sleep over.

So of the 3 teams with so much variability that its difficult to put a numbers to them ... I really don't think any of the 3 are the "strongest teams out there".