| | You are not authorized to post a reply. |
| | Author | Messages | |
2005 D&D Miniatures Champion
Fenris
Sergeant
974 Posts
    
 | | 06/28/2006 2:05 PM |
| I'm trying to figure out what a good "spawn curve" might be -- how many creatures of each cost should you include in a band, generally? I'm approching the problem by trying to figure out how many points, on average, you can expect each round from the dice. Not sure if that's a good approach, and I'm not terribly good at math, so feel free to point out what I'm missing.
Dreamblade uses 2d6 to determine spawn points, and 2d6 would normally give a bell curve, with 7 being more likely than any other number. But when you factor in 0 spawn points for any combination of rolls that includes a 1, that changes things. Suddenly zero is actually the most common result, with 11 combinations containing a 1. Next most common result is 8 spawn points, with 5 combinations totalling to 8. I graphed the results, take a look:
I don't really know what to make of that, other than to just pretend the spike at the front doesn't exist, and build a curve based on the latter part of the graph.
Just taking an average of all the possible combinations, including all the rolls that result in zero spawn points, gives a result of 5.55. This suggests that your spawn curve shouldn't center on 8, like the chart above might otherwise indicate.
Then there's the whole issue of getting spawn points for dead creatures, and through various other abilities. And the fact that consecutive turns where a 1 is rolled result in a re-roll. Hrm.
Math hurt Hulk's brain! Hulk smash puny minis! [)]
(No, but seriously, how do you calculate a decent spawn curve?) | |
Read my championship tournament report.
Successful trades:
xBrendanx | Cha0tic G0od | ThatOneGuy | BrazenWood | Finley | Arris | Drift x2 | Neonmage | Cthulu's Librarian | Kalvos | PatEllis15 | Paradox1995 | Rhane | Gildogg | Darthrau | Vrecknidj x2 | erian_7 | Octavius Drowslayer x2 | stheis | Garate | thenameless | Lucky_ksu | ckissee | Corim Danex | Schooly_D | Username | |
|
Balduran I
Sergeant
404 Posts
| | 06/28/2006 2:12 PM |
| The 5.55 should be ignored, IMO, in favor of the 8. Reason being that on the 0 turns one cannot spawn at all, making it irrelevant if you've centered things on 5 or 6. When you are able to spawn, the value you'll get (barring mods) will center on 8.
Dunno math wise what such a binary thing is called (if on then what do I need, if off I need nothing), but it strikes me as more useful than the average. | |
| |
|
striderlotr
Commander
3370 Posts
| | 06/28/2006 2:30 PM |
| Also remember that you can spawn multiple minis in a turn, so having an average of 8 might mean you will get more creatures instead of a big one.
The other thing to remember is aspect cost. Every mini has an aspect cost and some require 2 or 3 of a type to bring that creature into play. If you do not have one of those aspect creatures in play or in the graveyard, then you need to add one for each aspect cost. A piece of that that Mike explained was that if on turn one you bring in a 2 point creature with 1 aspect is costs 3 spawn points. If you then bring out a 3 point creature with one of the ssme aspect cost, it only costs 3... as the aspect of the first on satifies the aspect cost of the second one.
Now I'm intrigued and want to play some more. | |
Sean Banks
Champion of Elementals
Official Organizer
Gen Con 05 maxminis Event | Winterfantasy 06 maxminis Event | Gen Con 06 maxminis Event | Winterfantasy 07 Community Event | |
| 
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| |
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| | 2005 D&D Miniatures Champion
Fenris
Sergeant
974 Posts
| | 06/28/2006 5:02 PM |
| Right, I understand how multiple-aspects in the spawn cost, or other spawn point abilities might change the baseline, but isn't it good to have a baseline first? It strikes me that it'd be useful to have something to work from, a generic rule of thumb that you should have X creatures of cost Y in your band.
I guess I'd be inclined to start from your suggestion Balduran, and just ignore the average due to the binary nature of spawn/no spawn. So, if we follow the bell-ish curve and ignore the spike at the beginning of the graph, that becomes something like:
3 of cost 8
2 each, cost 7, 6, 9, 10
1 each, cost 5, 4, 3, 11, 12
That looks too high to me. Maybe it's not, but it sure looks like there are too many expensive creatures in there.
| |
Read my championship tournament report.
Successful trades:
xBrendanx | Cha0tic G0od | ThatOneGuy | BrazenWood | Finley | Arris | Drift x2 | Neonmage | Cthulu's Librarian | Kalvos | PatEllis15 | Paradox1995 | Rhane | Gildogg | Darthrau | Vrecknidj x2 | erian_7 | Octavius Drowslayer x2 | stheis | Garate | thenameless | Lucky_ksu | ckissee | Corim Danex | Schooly_D | Username | |
|
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| |
Dagni
Sergeant
870 Posts
| | 06/28/2006 7:28 PM |
| Getting 0 spawn points matters only for guessing how many turns you will have with a spawn point round. Add in the tournament rule of rerolling 1s if there was no spawn the prior turn, and you end up with an average of 1 round in 4 skipping spawn. My calculations guess that a short game with few tied turns will have around 9 spawn rounds, and a long game with a lot of tied turns will have around 13 spawn rounds.
The issue of getting bonus spawn from dead creatures is also fairly simple, in terms of total number of spawn points desired, if not in terms of ideal spawn curve. By the time you clear out your reserves, guess the number of creatures that your band will, on average, still have alive. Conservatively, most bands will have at least 4 still alive, and probably all of it's locations. If all the other 12 creatures had died by endgame, that'd give you 24 bonus spawn points, in addition to the 9+ rounds of 8 spawn points average.
Therefore, from that, I'd guess that a typical band should have a good 96+ total spawn cost. Since relatively few games will give more than another 32 points of spawn, for an estimated high point of 128 spawn, anything higher than that probably won't make use of all their creatures. The curve also matters a lot, but if your curve is way over 100 total spawn cost, you can cheat some by putting in more low cost creatures, since they can be used in numerous combinations to satisfy most any single turn spawn point total.
The curve that you thought looked too high adds up to 123. Although that's under the ceiling I estimated, my guess from playing is that my estimate was pretty high to begin with, and constructed bands will use minis with spawn activation costs or Regenerating minis or other means to use extra spawn in long games with surprisingly high spawn totals. So yeah, that curve is probably a bit high.
Also, there's the question of expected consequence if actual results don't align with the curve. If the results are a little high, what happens is you are forced to waste a single point a few times. If the results are a little low, you run out of low cost minis, and waste the entire roll! So surely the 'ideal' curve won't match up *exactly* one-for-one with the bell curve of likely results; it should be just a shade under. Or, as has already been mentioned, use a 4+5 as one of it's '9's, etc.
- Dagni
| |
Proud member of the GRUUMSH fan club! | |
| DDM Constructed Champion 2006
derry
Warrior
268 Posts
| | 06/28/2006 8:20 PM |
| Robert (Dagni) brought up the point that you get spawn points for your creatures killed since last spawn. You also use spawn points for the abilities of certain creatures.
I think that bands without any creatures using spawn points for special abilities are easier to calculate for, but still not that easy. The mix and total spawn costs, for single and multiple aspects make for very interesting choices.
You reveal a lot by showing your warband and creature costs to your opponent at the beginning of the match. Your planned growth curve and some of the spawn strategies you plan to use during the match is can be figured out right away by your opponent (although at the beginning most opponents will be too busy trying to figure out what they are doing, not what you could do.)
Not only do creatures with spawn point special abilities make calculating the appropriate mix and total spawn costs for your band, they provide a very interesting way to both conceal your plans from your opponent, while simultaneously dynamically varying your actual spawns based on the needs of the match at that point. You can use or not use special abilities now in order to spawn less or more, depending on your anticipated needs. | |
Member of the Low Post Count but High Post Content Society | |
| 2005 D&D Miniatures Champion
Fenris
Sergeant
974 Posts
| | 06/29/2006 1:12 PM |
| Derry, Chris, Strider: I guess I could sum up your posts as saying "It's not that simple, there is no generic ideal curve, it will vary from band to band." I'm sure you're right, but I think I'm several steps behind you guys as far as understanding the complexity of the game. I also haven't seen the full set, so I'm still groping at the basics.
Further (and I should have mentioned this) I'm mostly concerned with sealed at this point, since that's what the big tournament at Gencon will be. In a sealed band, you probably won't have a lot of elaborate synergies, nor many multiples of units with abilities that use (or grant) spawn points.
I guess I'm pretty much trying to figure out how many units of each cost should be in a non-optimized, sealed band -- and I figure it'll be easier to adjust to optimized constructed bands once I have an understanding of that baseline.
Then again, you don't take out that many units from your sealed boosters anyway, so maybe making choices based on spawn cost isn't as good a strategy as some other method. *sigh* Only about a month till Gencon, how am I going to have time to practice both DDM and Dreamblade? {:(]
Dagni: I like your method of calculating total spawn cost of the entire band. I'll mull that over some. | |
Read my championship tournament report.
Successful trades:
xBrendanx | Cha0tic G0od | ThatOneGuy | BrazenWood | Finley | Arris | Drift x2 | Neonmage | Cthulu's Librarian | Kalvos | PatEllis15 | Paradox1995 | Rhane | Gildogg | Darthrau | Vrecknidj x2 | erian_7 | Octavius Drowslayer x2 | stheis | Garate | thenameless | Lucky_ksu | ckissee | Corim Danex | Schooly_D | Username | |
|
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| |
Amy
Skirmisher
7 Posts
| | 06/29/2006 2:49 PM |
| In a way, I think creating a "good" spawn curve in limited is easier than in constructed. My only experience in sealed is from the games I have played on VASSAL with the incomplete generator, but I think that still gives me a good idea of how to make it work well. First I sort the minis by aspect and spawn cost. Right away you can usually eliminate either an entire aspect(with only a few creatures) or at least a few miniatures that you know won't work well in your warband (Defenders with no advance, triple aspect cost pieces (like the Night Queen) without single aspect pieces for support...) After that the spawn curve actually helps me to eliminate or keep pieces. I try to keep each point value from 3-9 filled with at least one miniature if possible. If there are no 5s, I would try to compensate with a 4 or 6. Usually if I have more than 2 pieces with a cost of 10-12 I will try to get rid of them(unless you have some good ways of generating large amounts of bonus spawn). After all of that, I usually like to have more smaller creatures (2-6) to allow more flexibility in spawning.
I hope that people don't underestimate the importance of having a good spawn curve. I have won MANY games by being able to spawn more effectively than my opponent. Wasting a few spawn points isn't an automatic game-loss, but even a few points can put you a little behind. Being unable to spawn because you run out of creatures or can't spawn all of you expensive pieces with low rolls can severely hurt you in a game.
That all said, in limited you have to work with what you are given. I'm sure I will try hard to have a good spawn curve, but I will certainly sacrifice it some to ensure I have good combinations of miniatures and abilities. | |
| |
| DDM Constructed Champion 2006
derry
Warrior
268 Posts
| | 06/29/2006 5:32 PM |
| I concur with Amy. For sealed play your band building challenges are less complicated. Needing 16 pieces for a band, and having 23 minis with a starter and a booster (non counting any legal promos) means you may want to focus on the seven you will remove. Focusing on the seven you will remove allows you to reduce the number of decisions to make. For Dreamblade Day, I put all my minis in spawn cost order by aspect to quickly see what I had.
Exclusion Factors
1. Lineages: I had no lineages to factor in, but if I did I might have removed those minis costed for good lineage abilities that I could not use.
2. Smallest Aspect: Reguardless of the mix, probably one entire aspect is poorly represented in your 23 minis and could be entirely targeted for removeal. Almost half my minis were fear (as were both my rares), so madness had very few minis and I excluded all 4 of them.
3. Synergies: While looking for synnergies that you can benefit from, you can also see powers on some creatures that we be difficult to use. An example is without the movement capabilities built into the valor aspect, minis like the Ύ power mini with charge] can easily get locked in combat and eventually eliminated.
4. High Aspect Cost: Minis with high aspect cost are harder to put into play in a warband with three or more aspects. If the Aspect cost is paid with spawn points (after the first one), then those creatures are less efficient in practice. Paying more for the same mini hurts your competitive chances in matches.
[EDIT: removed possible spoiler] | |
Member of the Low Post Count but High Post Content Society | |
|
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| | 06/29/2006 5:51 PM |
| That piece is already spoiled; you're safe on that one.
In terms of a sealed plan (and this topic has strayed there) I did the same thing as derry during play. I eliminated my smallest faction, then eliminated figures that had synergies on which I could not capitalize. At that point I had five or six of my 23 eliminated and only had to chose one single additional figure to remove. I looked at the liklihood I would be able to use a specific ability and removed the most difficult as that would be the least efficient for its spawn points.
One thing of note - some of the minis will be weak in constructed where you'd face hyperoptimized combos - but could be just fine in limited ... the opposite case of what derry described (mini that combos well in constructed but could be missing key pieces in limited). Identify those as part of your analysis, pick your 16 figures and have at it! | |
Triangle DDM Skirmish Group | My Email | 45-ish trades and counting | Stuff for Trade
* * * Show your brother some love and click here * * * | |
|
Dagni
Sergeant
870 Posts
| | 06/29/2006 9:10 PM |
| quote:
Originally posted by Fenris
Derry, Chris, Strider: I guess I could sum up your posts as saying "It's not that simple, there is no generic ideal curve, it will vary from band to band." I'm sure you're right, but I think I'm several steps behind you guys as far as understanding the complexity of the game. I also haven't seen the full set, so I'm still groping at the basics.
Well, it might not be that simple... but that wasn't what I was trying to say. I think there's something like an ideal curve.
In part, I was just throwing out ideas and tips I had, and in part I was trying to say that particular curve was too high, and a bit too inflexible. By taking out a couple of high cost minis, and replacing with low cost ones, you gain much needed redundancy, and the ability to spawn more than one mini during some of your turns. Remember, the guideline is there won't likely be more than 13 spawn phases, so you should be safe spawning more than one mini on a few turns, preferably including the first turn.
Or, you could just take about that same curve, and lower the cost of every mini by one. Again, that would seem a bit more reasonable. Also, for Sealed, with a few games of Simulated sealed you'll find not so much the curve you're necessarily aiming for as much as the curve of what you'll get. There are a lot of 5 and 6 cost commons, and only a couple 8 cost commons, for example.
That's not to say you shouldn't tweak what you take out based on your curve, but from playing games I've found having several 5s and 6s and having zero 8s can work out alright if the rest of the curve is filled out well. It's normal for me to have zero 8s, but 2 each of 7 and 9, for example. The 2 8 cost commons, by the way, aren't sure plays, each having a disadvantage, which is why I've had zero 8-drops in many Sealed games. In comparison, I like several of the 9 cost commons and uncommons, and seem to always have 2 or even 3 in my band.
- Dagni | |
Proud member of the GRUUMSH fan club! | |
| 2005 D&D Miniatures Champion
Fenris
Sergeant
974 Posts
| | 06/30/2006 11:01 AM |
| quote:
Originally posted by Dagni
Well, it might not be that simple... but that wasn't what I was trying to say. I think there's something like an ideal curve.
In part, I was just throwing out ideas and tips I had, and in part I was trying to say that particular curve was too high, and a bit too inflexible.
[snip]
That's not to say you shouldn't tweak what you take out based on your curve, but from playing games I've found having several 5s and 6s and having zero 8s can work out alright if the rest of the curve is filled out well. It's normal for me to have zero 8s, but 2 each of 7 and 9, for example. The 2 8 cost commons, by the way, aren't sure plays, each having a disadvantage, which is why I've had zero 8-drops in many Sealed games. In comparison, I like several of the 9 cost commons and uncommons, and seem to always have 2 or even 3 in my band.
Right, gotcha, I did like the way you were approaching it from the angle of total spawn cost of the entire band. I hadn't thought of looking at it that way. [:)]
I wonder about not having any 8s, do we know for sure that there are only two commons with cost 8 in the set? How many 7s and 9s I wonder. It'd be interesting, tho maybe not terribly useful, to plot qty vs cost on a graph too, and see if it aligns at all.
With a curve skewed toward the high end, say maybe six minis of cost 9 or more, I'd imagine it's more important to keep it down to 3 or 2 aspects. Hmm.
I need to play more games. | |
Read my championship tournament report.
Successful trades:
xBrendanx | Cha0tic G0od | ThatOneGuy | BrazenWood | Finley | Arris | Drift x2 | Neonmage | Cthulu's Librarian | Kalvos | PatEllis15 | Paradox1995 | Rhane | Gildogg | Darthrau | Vrecknidj x2 | erian_7 | Octavius Drowslayer x2 | stheis | Garate | thenameless | Lucky_ksu | ckissee | Corim Danex | Schooly_D | Username | |
| 2005 D&D Miniatures Champion
Fenris
Sergeant
974 Posts
| | 06/30/2006 11:04 AM |
| | Oh, and thanks Amy and Derry and Chris, those last few posts are very helpful. [:)] | |
Read my championship tournament report.
Successful trades:
xBrendanx | Cha0tic G0od | ThatOneGuy | BrazenWood | Finley | Arris | Drift x2 | Neonmage | Cthulu's Librarian | Kalvos | PatEllis15 | Paradox1995 | Rhane | Gildogg | Darthrau | Vrecknidj x2 | erian_7 | Octavius Drowslayer x2 | stheis | Garate | thenameless | Lucky_ksu | ckissee | Corim Danex | Schooly_D | Username | |
| DDM Constructed Champion 2006
derry
Warrior
268 Posts
| | 06/30/2006 12:02 PM |
| A constructed warband I borrowed to play had a Noble Dragon (ND) with a spawn cost of 12V. When looking at the warband prior to the match I was concerned about my ability to spawn the creature, but during the match we had a no spawn turn, and I lost three creatures over those two turns. So I was able to spawn the ND because we rolled a combined 9 (+6 = 15 spawn points to spend.)
But, I really wanted to spawn other creatures that spawn phase, but I was sort of forced to bring in the ND since I might not ever get the chance again (and I was right as it turned out.) The ND is not what I needed at that point in the match and it got killed that same turn (hot passion creatures just blew it away and I got no Payback since I rolled no blades .)
I agree with Robert in that there is a spawn curve can be calculated and used to build a warband. I think that understanding the math behind the curve for both your own and your opponent's warband is important. In the beginning I think just being aware that their is a curve and having a basic understanding of the tradeoffs that are made with warbands at various levels is fine. As strategy and play evolves, I can definitely see players doing a much better job constructing warbands matching a better practical curve that correlates with their play style and the needs of the warband.
I think you can mitigate miscalculation (especially those on the low side) by putting in creatures with Regenerate (and somewhat those with Reincarnate.) But you never know what the next match spawn distribution and total will be. In the matches I have played so far, even using the tournament rules preventing two no spawn turns in a row, I have seen low total spawn games (~70) and high (~140+). (I am not sure 140 is high, but it is much higher than 70.)
Note: I will get the exact numbers later, my logbook is at a game store with the minis right now. | |
Member of the Low Post Count but High Post Content Society | |
|
Balduran I
Sergeant
404 Posts
| | 06/30/2006 1:21 PM |
| One thing that struck me playing is this:
Running a constructed warband well will demand knowing exactly what you're going to spawn on turn 1 before the dice hit the table. The better player will have his minis pre-picked for every number from 5 through 11. Even multiple picks, changing depending on what you're facing.
It may even evolve that the better players have multiple turn's spawns picked well in advance. It's not like Magic, where you don't know what'll be available in your hand. There's a finite number of spawn totals, and you know what you have in your reserves. Because of that there can be considerable planning done. | |
| |
|
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| |
jgsugden
Commander
4320 Posts
Walnut Creek, CA
| | 06/30/2006 5:50 PM |
| To make a long story short, I think the best bands don't focus on the probability curve, at least not in the beginning. They focus on gaining an early advantage and pushing that advantage until the end of the game. Thus, they must be able to efficiently handle all early game combinations and be able to add pressure until the game ends without running out of resources.
The key, to me, is focusing on the first two spawnings. You need to have a game plan for using all 81 combinations of spawn points you're likely to get in the first 2 spawnings (assuming no bonus spawn points are gathered before the second spawning - something that I think is a fairly reasonable assumption in most cases - and even when it is not, something that will likely not hurt you as rolls of 11 or 12 on the second roll after 1 creature dies are going to be pretty rare, and will probably be easy to adjust to handle int he early game).
Those 2 round plans need to give you:- Enough pieces to perform various tasks on the Dreamscape.
- Be internally synergistic between pieces.
- Facilitate affinity in the later game.
- Use the least number of figures possible that will not be useful in mid and late game strategies.
Ideally, you'll be able to use no more than 8 total figures in these early game plans. That will leave you with 8 figures for planning for the rest of the game.
This is where the die rolling curve comes into play. However, I do not think it is wise to focus on the number 8 shown in Fenris' chart, as this is the time when you'll start seeing bonus points for destroyed creatures. Instead, I think you need to stretch the curve and assume that some bonus points will appear at some time and will not be rpesent at other times. This allows you to continue to plan for options that cover a wide range of die rolls... from 6 to 14. You only need a few options at the extremes of this range, but you want to have a number of options in the 6 to 10 range using 1 or 2 pieces to achieve those numbers... probably averaging closer to 2, as there will be many turns in which no creatures are spawned. As you'll only have ~8 figures to try to cover this range, and you need to trry to cover possible die rolls for many turns, there will be gaps. However if you focus on one or two combinations that ciover the edges of the probability curve and a lot of combinations that get at the core of it, you should be able to get yourself some pretty good option coverage.
This planning is difficult in one sense because we won't use up all the pieces involved in our initial plans during those first 2 rounds. Only 1 of our 81 possible options will actually pop up. Accordingly, any figures in any of those other plans that are not included in the actual plan used we used will be X-factors.
Screw those X-factors. Plan the rest of your pieces as if those X-factors didn't exist. This should help guarantee options for you in the mid-game where you'll need to continue to press any advantage you've gained.
Returning to the X-factor pieces left over from your early strategy, those X-factor pieces should be cheaper pieces, so you'll be able to toss them out towards the end of the match after you've efficiently implemented your mid-game strategy.
This is a generalized approach, but it would be how I start my building process in constructed:
Specific plans for the first 81 options in the first 2 spawnings using the fewest number of figures possible, then generalized plans based around an elevated spawn point curve range for the remaining pieces.
What does this get me in total as a building plan?
Early game pieces (1st 2 spawnings): Plan out 81 different combinations of efficient piece spawning using as few figures as possible, but trying to put 3 or (hopefully) 4 pieces on the board by the end of the second spawning. Make these plans using the fewest total number of pieces possible.
Mid game: (3rd - 8th? spawning): Fill out the remainder of your 16 figure warband with combinations of creatures built to facilitate the elevated die curve while spawning 1 or 2 creatuers per spawn and based upon the affiliations facilitated by the early game strategy.
Late game: You can't really plan this far in advance, because so many different things can happen in the early and mid-game that any late game plans are meaningless. In addition, the game advanatage has already been established (in most games) before the late game begins. Shifting the advantage in the game at this late juncture is very difficult. Accordingly, don't plan for it... just make do with whatever you have left over.
How does that sound? | |
Champion of Meepo _*_ Myztek on the Wizards Boards. _*_ (2206 DDM on 03/06/06)
Please note: The use of the  indicates an attempt at humor ... often a bad attempt.
BAD EBAY SELLERS LIST (CLICK HERE): AVOID AT ALL COSTS
| |
|
| | You are not authorized to post a reply. |
|
| |
ActiveForums 3.7 | You must be signed in to participate in the
games. |