THE HOSER REPORT

	BAS	STA	SLA	PST	LUA	PLU	WST	TPS	AIM	PRP
BA	M	W	W	W	U	U	W	U	U	U
HA	U	W	W	W	U	U	W	U	W	W
MS	W	W	U	M	U	U	W	U	U	U
WF	W	W	U	M	U	U	W	U	U	U
WH	W	W	U	W	U	U	U	M	W	U
MA	W	W	U	M	U	U	U	M	W	U
SS	U	W	U	W	W	W	U	W	W	W
LO	U	W	W	W	W	W	U	W	W	W
BS	M	W	W	W	U	W	W	W	M	W
EP	U	W	W	W	W	W	U	W	W	W
SH	U	W	W	W	W	W	U	W	W	W
DA	U	W	W	W	W	W	W	U	W	U
SC	U	W	W	W	M	W	W	W	W	W
GA	W	W	W	U	U	U	W	U	U	U
HL	W	W	U	W	M	M	U	U	W	U
ML	W	W	U	U	U	U	U	U	U	U
QS	W	W	U	W	U	U	W	W	W	U
LS	U	W	U	W	W	W	U	W	W	W
GS	W	W	M	W	U	U	W	W	W	U
FI	M	U	U	U	U	U	U	U	W	U

W = Well suited, M = Marginal, U - Unorthodox, underlined cells are uncertain.

Enough of that. On to fighting styles. Some people have suggested that the famed `luck factor' might make trying to relate style, initial attributes, and skills to expert unreliable. Maybe so. I've got some interesting thoughts on the luck factor latter in the newsletter. However, during character design I still think it would be useful to know what attributes to dump those last points into to maximize a specific skill area. We've seen lots of PRP figures, here are some for the PLU (next issue we'll look at some LUA). ST CN SZ WT WL SP DF Hand Original: 15 7 10 15 16 6 15 R Final: 15 7 10 15 16 6 15 Increases: Parry + 8 or 9 (skills) = expert (learned two same fight declared expert) ST CN SZ WT WL SP DF Hand Original: 11 11 10 19 9 11 13 L Final: 12 11 10 19 13 11 13 Increases: Attack +6 = expert, initiative +7 = expert, Parry +9 = expert ST CN SZ WT WL SP DF Hand Original: 11 9 10 17 15 5 17 R Final: 12 11 10 17 16 5 17 Increases: Attack +2 = expert, Defense +8 = expert, Initiative +4 = expert, Parry + 9 = expert ST CN SZ WT WL SP DF Hand Original: 11 17 9 11 13 15 10 R Final: 13 19 9 13 15 16 11 Increases: Attack +9, WT,WL,DF + 1 = expert, Defense + 13, WT,DF +1 = expert, Initiative +6, WT + 1 = expert, Parry + 12, WT,WL,DF + 1 = expert ST CN SZ WT WL SP DF Hand Original: 14 5 6 17 17 6 19 R Final: 14 7 6 17 18 6 19 Increases: Attack + 3 = advanced expert, initiative + 5 = expert, Parry +7 = expert, Riposte + 7 = expert

HOSE KNOWS Questions and Answers

Q: In issue #3, Mark Ferris referred to a 13 WL equaling to a 65% chance to raise a stat the first time. How about some further information on the relationship between WL and the % chance to raise stats the first time, second time, third, etc. A: The common wisdom holds that stat raises are based on a d20. Since WL is approximately 3-20, every point or it adds 5% (100/20) to your base chance the first time. If successful, the next raise has about 1/2 the chance of happening. Say your gladiator has a 19 WL. If you attempt to raise SP (for example) for the first time, you have a 95% chance for success (19 * 5). The next time you make the attempt, your chance is half of that, or 47.5% (its probably safe to round to the nearest 5%). The third attempt is at half of that, or, (rounding up) 25%. Failed attempts change nothing. In AD, the first 3 raises are treated as the first. Using the above example, you have a 95% chance for EACH of the first 3 raises (remember, misses won't affect the odds). Then, the 4th raise is treated at 1/2, the 5th at 1/4, etc. As you can see, with a 10 WL you only have an initial (and best) chance of 50% per attribute. Another good reason why there will never be a dominant character with a WL of less than 10 (I'll catch some flak for that statement I'm sure). Lastly, the game designer had (at one time) the intent that a character would always have a 10 or 15% minimum chance of success. Q: You're supposed to have a better chance to learn skills against a more experienced warrior, if you win, if it's a long fight, etc. My experience has been very mixed in this regard. Your comments please on this. Also, how many skills per attempt would a particular WT level expect to learn? A: It sounds like you're already aware of the factors that increase learning. The thing to remember, not only in regards to learning but performing any other action, is that this is a game based on chance. There are no guarantees of doing, anything, including getting x number of skills per fight. Your character blows a die roll inside the computer, and you're hosed regardless of how good you think he is. Lots of managers grumble about program bugs, the moderators, etc. when luck goes against them. They would be better off writing the fight off as bad luck and moving on. As far as determining skills per turn based on WT alone, forget it. I have a BAS with a 13 WT (13 WL) that can pull down 5 skills per turn. On the other hand, I have a LUA with a 17 WT (17 WL) that is lucky to get 3 per turn. The best thing I can tell you here is the more WT the better. Sorry. Q: I just recieved a replacement character. His stats are: ST CN SZ WT WL SP DF 10 20 12 8 5 7 8 I'm new to Duelmasters and am not sure if this roll up has potential or not... I was thinking of making him a LUA with: ST CN SZ WT WL SP DF 13 20 12 13 11 7 8 (raise to 9 1st fight) I realize that he would probably be slow and uncoordinated but with CN 20 and ST 13 he could probably take any punishment thrown at him and still wear down his opponent. What do you think? A: I never worry too much about the uncoordinated statement. My first reaction was DA, but that LUA you have doesn't look that bad. I don't feel that this is a long term tharacter. With that in mind, an 11 WT would be sufficient. Use the extra points to get DF as high as possible (11 if you switch a point from ST also). He should win maybe 5 of his first 7. Then send him DA and hope for a better replacement. This should help your W/L quite a bit if you have a new team. Q: Do you have any advice on LUA? I've had 5 killed. Usually around fight 6. The death rate seems fairly high for the offensive styles. A: I don't think that the death rate is very high for anyone. However, offensive styles tend to use up their END rather quickly, reaching exhaustion in the first 2 or 3 minutes. When you're exhausted only 3 things can happen and all of them are bad. Hopefully the 5 you lost were all 6-0 or close. This is ideal because not only do you get that tremendous boost to the W/L, but you get to own someone for 4 turns via the blood feud (you should have someone on your roster who can get at least 3 wins out of it). In general, use mid to high offenive effort (for attack percentages) with mid level activity (to conserve END as best possible) vs. defensive fighters. Its safe to assume that you will always have determination via initiative. Against aggressive fighters, don't hold back anything. Vs. aggressive opponents the fight will almost always be over in two minutes (usually less) so END is not a big concern. Even if he you can dodge well, its not your strength. Use light armor (ALE, F), and try a LO if you can use it. The ENC is 1/2 that of the LS, but you don't have to hit a given opponent twice as much to end the fight. If you like to take chances, you can forget a back up weapon (keeping ENC down).

THE DUELMASTERS HANDBOOK & THE LUCK FACTOR

Last week a very interesting item came across my desk, the Duelmasters handbook by Mike La Plante. Its a pretty slick little rag, 30 pages spiral bound. Well thought out, well put together. Unfortunately, at this time it is apparently not being sold and is only in limited distribution. The contents include: Stats Vs. skills & capabilities, formulas for damage, hit points, endurance, encumbrance, coordination, intelligence, weapons table with damage ratings, weapons vs. stats & style, recommended set ups and strategies for each style, and more. With updates and more to come.. The Duelmasters Handbook is the best attempt I have seen to date to quantify the game. I'm sure that most of us are familiar with games like D&D where the player can go to various tables to get a numerical value for such things as hit points, encumbrance, etc. As we all know, in Duelmasters RSI is sitting on this type of information like a mother hen. This handbook contains tons of tables and practical information (linear regression is used to derive much of it). The formulas are likely not the exact formulas in use by the moderators, but are very workable. I don't neccesarily agree with everything in it, but it will give all players new insights into the game (with any luck the authors can be persuaded to sell it). It has been put together over the last 3 years by several managers and apparantly sources close to RSI employees (leaks are not uncommon, I understand lots of sensitive Hyborian info is floating around the LA area). The authors have given me permission to use items from the Duelmasters Handbook in the HR. Stay tuned. Among the many things that give food for thought is the idea of the "luck factor" (LUF). This is to be distinguished from the infamous "will roll" which is akin to the D&D morale check. The LUF is used to explain many things as we all know. In fact, it seems to come into play everywhere (at least, everyhere something odd needs explaining!). Lets take a closer look at the luck factor. Now, one of the game premises is that as a warrior is trained (learns skills) it increases ones mastery or probabillty of performing certain actions. The common wisdom holds that learning a skill (say attack, my favorite) adds "x" percent to ones base chance of hitting the opponent. If you have this luck factor popping up everywhere, or if this factor was a large percent of the act in question, the result would be an unpredictable game. In other words, if the program goes through an elaborate procedure of taking your base attack percent and modifying it (for weapon, tactic, offensive effort, encumbrance, exhaustion, attack location, and anything else which may effect the outcome), then adds a 50% chance of "luck", what value is the modified skill level in the first place? When a gladiator who has learned 20 skills fights one that has learned only 10, we are all pretty sure of the outcome. Fighters from the bottom of the rankings do not regularly defeat those at the top. The luck factor can't be that large. Let me digress for a moment to one of my last visits to RSI's underground factory. I was watching the programmer Chuck Kraver input several replacement characters (one of which was mine). As I have said in the past, if you ever get a chance to visit RSI or go to the Tourney, watching the computer screen as things are happening is a must. Getting back to the story, the character overview program was doing its thing. Checking to see that the point total was equal to 84. Checking to see if this was in fact a legitimate replacement (we've all heard about the counterfiets - I heard of one manager who actually had the nerve to submit an entire team of SZ 3 characters!). Then, the number 5, 10, or 15 was assigned to each of the new characters. I asked Chuck what the significance of this was. He replied that it was a rating or evaluation devised by the game designer to measure the soundness of a character design. It sounded bogus to me, but at the time it was clear that I would not get a straight answer and it didn't strike me as important. It was forgotten. Until a few days ago. After receiving the Handbook I called the authors (my next phone bill will have a 10 kill desire) and the ensuing conversation turned to skills and the "luck factor". Mike and Mark La Plante said that each fighter had a bonus or up to 3 skills during roll up, making comparison of simflar (even identical) fighter difficult. And, the LUF was randomly assigned to each character, being 5, 10, or 15 (0 is also possible). If I had not personally seen the roll up program in action, I would have likely dismissed this. But, I had seen it on RSI's computer screen. I contacted a few other managers who don't communicate with each other, and got the same account. Hmmm. Divide 5, 10, and 15 by 1, 2, and 3, and you get 5 each time. This is a number that has been referred to as the percentage value of a single skill by many people. Managers of very experienced fighters will verify that the upper limit of learning in any one area is about 20 skills (less if attributes affecting the skill in question have been raised before the 20th skill). 20 * 5 = 100. As in 100 percent. It fits too nicely. Does this mean that if a character has learned 20 attack skills he has a 100% chance of making a successful attack (i.e., never makes a "wide" or "wild" attack). Not likely. Why? It would mean that a rookie fighter (or one that has never learned an attack skill) would have a 0% chance of making a successful attack. That is clearly not the case. Obviously every (or most every) fighter starts above 0%. Which implies that you can exceed 100% (if the above assumptions hold). Also, it appears that the luck factor is no larger than 15%. This seems not too large as to throw results, yet large enough to explain the odd things we notice from turn to turn. Now on to the number to expert (NEX) or more rarely number to advanced expert (NAD) case. Suppose that you have two new characters with identical stats, and the same fighting style. In the absence of the LUF you would expect that each would require the same number of skills to reach expert status in a given area. If the LUF as described above opetates here, there could be a variance of up to 3 skills in the NEX (similar to a confidence band in linear regression excepting it is not plus or minus). To illustrate lets plot WT vs. NEX attack for the case of no luck factor (A) and luck factor (B) assuming all other factors are equal. (Really cheesy graph here) The point I am trying to make is that ST, WL, DF, and style being equal, every odd point of WT should reduce the NEX for attack skills. But how much? Also you dont know what LUF will be assigned to the new recruit. However, if you get off your duff and send your NEX info in for publication (anonymously if you prefer - plain brown wrappers are ok), eventually the influence of each stat on each skill area will be clear as will the influence of each style. Then when you hit the first expert, you compare the actual to predicted and you know what the LUF for that fighter is. It should be an across the board type of thing. I'm out of room. More on this next time.

FEELS GREAT! LESS KILLING FEELS GREAT! LESS KILLING

In HR #4 I invited managers to voice their opinions on the issue of killing in the arenas. Here is a sampling of the responses so far. I have had my best warrior killed, 3 A. E. and 1 exp. in 21 fights. I realize that I am in a deadly arena (two teams are just about to reach 20 kills each In 40 turns; they're both in the top 5 so they know what they are doing) and personally I like it. It gives our arena character. It takes skill to kill and I believe credit should be given where credit is due. High kill desires are supposed to make a warrior less in control. This is a definate advantage to his opponent. If the opponent loses or dies, this just shows how much better the first fighter is. Again, I can't stress this enough, it takes skill to kill and I believe credit should be given where credit is due. Many people wouldn't buy that story, so here's what I tell them to put them in their places. This is Duelmasters! Let's face it, in a gladiatorial combat game, death is, as it should be, part of the game. Do you think the ancient Romans used rubber weapons? Like it or not, each manager must be wllling to deal with death. - Alan Yip ... As to the wild eyed managers and their sunning dogs going for the kill, also add their screaming canaries. Editor you will not believe this but there is even a team on the Isle. The Hose Machine that goes for the kill. Yes. True. The data `seems' to show that few arrive on the Isle with a lot of kills, although the figures can lie, (You) would have to factor in the large number of TPS. (It) would seem that RSI promotes kills to stir up 'fuzzin and feuding'. That is Ok. (There is) intimidation. Discourage challenges. Means more of your own work.........Kills might be fun, although I would rather try to figure out how to best a warrior who has beaten mine 10 times. That is a matter of taste. IS IT A PLUS? All things equal, who will win? The warrior with a 7 kill or a 1 kill? I don't know, I'd suspect the 1 kill. If we must have kills why not give them more class. Something besides "Hosefoe wails and drops dead". Something elegant. More Erol Flynn like. More dashing. An occasional "Runs him through". - Victor Melucci

ODDS & ENDS

Among some of the suggestions recieved recently was a suggestion to go to a longer format (maybe 7 or 8 pages). Since such a move would increase my costs significantly, this would involve a price increase. Another manager suggested generating sample replacement characters in each HR and inviting managers to design the character and explain their reasoning. I'm waiting for your input. It looks like, the "computer" Spy Report is gone. Not to many managers liked the compu-spy. I'm surprised it generated such strong opinions. Too bad for RSI, I'm sure it increased their productivity. From experience. I can tell you that writing a good report day after day for multiple arenas burns you out. Its very difflcult to keep up on things such as alliances, feuds, and what not. One astute manager noticed that in issue #5 1 raised my price back to the old level. Yes yes, fluctuating market conditions. A new definition for frustrating: Frustrating is when you learn the "favorite" offense and activity of one of your most promising gladiators, and he goes on a 4 fight losing streak when you implement the "favorites". Jeff Morgan