Thoughts on the German Scoring System

Good morning gamers,

As a preliminary note, the blog just passed the 2 million (reported) page views mark - from all of us here at TMAT, thanks for taking the time to read what we're posting! 

This past June, we here at TMAT ran our annual Grand Tournament and we decided to use the "German Scoring System" for awarding tournament points instead of our usual 2-to-10 system for major/minor wins/losses (more on this later). The German Scoring System was different and had some very interesting take-aways from our tournament. While there are certainly other players who could weigh in more on how valuable this system is, I wanted to present my review of the system based on the 16 games we had at our event and provide some observations about it. Let's dig in!

What Are The Distinctives of the German Scoring System?

To be pedantic, in most tournaments (especially those in America), players compare their victory points (VPs) and are awarded a victory or loss based on whether one player has more VPs than the other (a victory) or fewer VPs than the other (a loss). These are either "major wins/losses" if one player has double the number of VPs of the other or "minor wins/losses" if one player has more but not double. In many tournament settings, a major victory is only awarded if the winning player scores more than X number of VPs to encourage them to "actually play the scenario". I have recently been implemeting this in our tournaments, requiring 6 VPs to be scored to get a major victory. Without this limitation, a 2-1 victory is awarded the same number of Tournament Points (TPs) as a 7-2 victory or a 12-0 victory.

Depending on whether you get a major/minor victory/loss, you get a certain number of TPs. How many TPs you get for a major/minor win/loss or a draw depends on your tournament, but we here at TMAT have traditionally done 10 points for a major win, 8 points for a minor win, 6 points for a draw, 4 points for a minor loss, and 2 points for a major loss.

If you compare this to the MESBG Matched Play Guide, you'll find on page 5 that it's recommended that a win (any margin) is awarded 3 points, a draw is awarded 1 point, and a loss gets you no points. VPs are used to break ties (first by VP differential, then by total VPs scored). Further tie breakers are recommended here as well, including most enemy army leaders killed, number of secret objectives completed, or favorite opponent rankings. This is different from what we do, mostly because I feel it has less granularity (and because ending your day with 0 TPs sounds cruel to me).

With the German Scoring System, however, there are 20 points up for grabs and they're allocated between the players based on how many Victory Points they receive in the round. Here's what the grid looks like (and for more information on this grid, check out the Unexpected Podcast episode on this subject - it's excellent):

As you can see, the scoring for this is going to be . . . huge. Very huge - especially if one side dominates and the other gets slammed. But what advantages are there for this kind of system? Well, let's take a look at some of the reasons that we found . . .

Why Do This?

The first advantage of this system is that it provides a mitigation for the "no win scenario" - sometimes in MESBG, you look across the table at your opponent's army, you shake your head, and you feel like saying, "Well, I can't win this one." So what do you do? Well, you say, "Great game - wanna get a drink?" Not much fun, if you ask me.

With the German Scoring System, you can look at the scenario and your opponent's list and say to yourself, "Well, I probably can't win this game of To The Death - he's got a crossbow gunline and I've got low Defense troops and no archery. BUT if I can keep my banner alive, kill his banner, and kill his army leader, I can get 5 VPs - and even if he shoots everyone except my banner guy off the table, I'll get a 10-5 loss at least, so let's shoot for that." Instead of saying, "Well, he's going to break me before I break him and get a major victory, what's the point?", we're instead asking the question, "How can I deny VPs to my opponent?"

This underscores the first important point of this system:

Almost every victory point you score helps you get more tournament points - and almost every victory point you deny gets you tournament points.

While there are situations when one side has scored less than 3 VPs that an extra point here or there doesn't actually do anything (the winner is still going to get 17 or 18 TPs), in general, if you can steal a VP here or there, you're probably getting a TP or two that would have gone to your opponent. There's always a reason to play the game as hard as you can. It also makes it so that each lost VP is painful - your overall score at the end of the tournament can be affected by a wounded army leader in one game or a banner you couldn't kill in another game. Breaking on the last turn can have HUGE consequences as well - every round of combat and every arrow fired can matter until you reach the end of the match.

The second thing about this scoring system is that to get maximum TPs, you have to get a 12-0 win, though getting a 6-0 win will get you 19 TPs, so there's room for generosity, I guess, if you can keep your opponent from getting any VPs. If this seems a little mean . . . well, that's because it is. If you can keep your opponent from getting 3 VPs during the course of the game and make sure you get at least 6 VPs, you're guaranteed to get 15-19 TPs - and if you can get the elusive 12-0 win, you can get 20 TPs. It can encourage some players to go for the jugular a bit, but if it's a competitive event, this scoring system will reward players greatly for denying their opponents any VPs. This leads to our second principle, though:

Unless you get a 12-0 loss, you're guaranteed to get at least some tournament points - and so long as you get at least 3 victory points, you'll get at least 4 tournament points for your efforts.

Four TPs out of a possible twenty isn't great, but it's something - and in some tournament settings (see the Matched Play Guide discussion above), a loss of any kind can award you 0 TPs - which means if you got an 8-4 major loss, those 4 VPs didn't do anything for you. By contrast, a German Scoring System would see that you gained 4 VPs and denied your opponent 4 VPs and would give 14 TPs to the winner and 6 TPs to the loser. Not bad.

Third, from first glance, the fact that small changes in VPs scored can change the number of TPs awarded to each player should make it so that ties in tournament points will be rare in tournaments using the German Scoring System. While this does appear to be true after lots of games have been played, after the first round or two, it's possible for players to have tied scores if you're playing the same scenario (where the potential for getting the same VP score is higher than if all tables are playing random scenarios) or if you're vetoing from a common scenario pool (where all the scenarios are similar and can award VPs in similar ways). This leads to our third principle about the German Scoring System:

Have a good tie breaker method in place, because when you need it, you need it to be good.

The fourth and final observation I have is that in most scenarios, there are lots of ways to score VPs - and as a result, there's a wide distribution of possible TP scores you can get. Some scenarios have only three scoring mechanics (the "point" of the scenario, breaking the enemy, and wounding the army leader), but in others, there are lots of categories that can give you points (whether it's multiple objectives, keeping banners alive, or killing/protecting certain models). We'll go into this deeper in our next section, but it bears mentioning here that because many scenarios have several categories for scoring points, there are lots of possible ways to get to an end score - and those possible end scores can lead to some very interesting TP results.

But the other side of the coin is true too: in some scenarios, the VPs can be heavily slanted to one or two categories (usually the ones that award 3-5-7 VPs for doing something more than your opponent), which means that it's easy to get a 7-0 win or a 6-1 win, which can give the winner 17+ TPs and leave the loser with 0-3 TPs. This brings us to our final principle:

If a tournament director wants players to have the greatest chance of having a close game, certain scenarios should be avoided. 

Let's look at this now . . .

So What Kind of Scores Are Reasonable to Expect from The German Scoring System?

While the chart maps out what the TP allocation would be for every combination of VPs from 0-0 to 12-12, each scenario has different possible scores on the table - you won't use everything on the chart. Realistically, you won't ever have a 12-12 draw since every scenario gives at least 1 VP for doing something your opponent didn't do - but at least the chart is consistent. To get a better view of what the possible scores are, let's look at an old favorite (or more appropriately, a scenario a lot of people hate): Reconnoitre.

Recon gives you victory points in three different categories:

• Breaking the enemy army: 1 VP if both players break the other, 3 VPs to one player if his opponent is broken and his own force is unbroken;
• Wounding the enemy army leader: 1 VP if the enemy army leader is wounded, 2 VPs if the enemy army leader is killed; and
• Getting models off the board: 3 VPs if you get 1+ models off the board and have more models off than your opponent, 5 VPs if you get 2+ models off the board and at least double your opponent's models, or 7 VPs if you get 3+ models off the board and at least triple your opponent's models.
  

These are actually very similar scoring rules to what you'll find in other scenarios, like Lords of Battle or Hold Ground - one category awards the majority of points for the scenario to one side only. Let's see how this affects the potential scoring results for the game.

Breaking the enemy is contingent on both players and results in four possible outcomes:

• 0 VPs awarded to both sides if neither force is broken (0-0);
• 1 VP awarded to both sides if both forces are broken (1-1); and
  
• 3 VPs to one side (or the other) if he breaks the other and is unbroken (3-0 or 0-3).
  

For each of these four potential scores, there are seven possible combinations for getting models across the opposing player's board edge:

• One side (or the other) gets more models off than the other, but not double (3-0 or 0-3);
• One side (or the other) gets double the models off than the other, but not triple (5-0 or 0-5);
• One side (or the other) gets triple the models off than the other (7-0 or 0-7); or
• Neither side gets models off the board (0-0).
  

This gives us 28 possible combinations (4 x 7) for the scoring of these two categories (ranging from a 0-0 score if both sides are unbroken and no one has gotten off the board to a 10-0 if one side has broken the enemy without being broken and gotten 3+ models off the board while 0-1 models have gotten off for the other side). When we look at the army leader scoring possibilities, we have nine possible outcomes:

• One side (but not the other) has wounded the enemy army leader (1-0 or 0-1);
• One side (but not the other) has killed the enemy army leader (2-0 or 0-2);
• One side (or the other) has killed the enemy army leader while the opposite side has only wounded their army leader (2-1 or 1-2);
• Both sides wound the enemy army leader but don't kill him (1-1);
• Both sides kill the enemy army leader (2-2); or
• Neither side wounds the enemy army leader (0-0).
  

If we have nine possible combinations for wounding/killing the enemy army leader for all 28 possible combinations of getting models off the board or breaking the enemy, that gives us 252 possible scoring outcomes. Who knew, right? 

When translated into the graph, we get a smaller subset of scores, which you can see here:

Click to view this in large mode - trying to get 170 scoring combos in a line made the font really small . . .

As you might be able to see here, there are 19 possible scoring combinations that have four distinct ways to get their value - and a bunch more than have 2-3 ways to get their value. These "most common scores" are:

• 1-5 (and 5-1) - 16 TPs for the winner
  
• 1-6 (and 6-1) - 17 TPs for the winner
  
• 1-7 (and 7-1) - 17 TPs for the winner
  
• 1-8 (and 8-1) - 17 TPs for the winner
  
• 2-5 (and 5-2) - 14 TPs for the winner
  
• 2-6 (and 6-2) - 15 TPs for the winner
  
• 2-7 (and 7-2) - 16 TPs for the winner
  
• 2-8 (and 8-2) - 16 TPs for the winner
  
• 3-5 (and 5-3) - 13 TPs for the winner
  
• 5-5 - 10 TPs for the winner
  

For most tournaments, most of the 1-X or 2-X results is going to award a major victory to the winning player - and the scores in the German system are pretty steep - but the losing player is still getting 3-10 TPs at the end of each game. 3 isn't great, but if he can just break the enemy and get a wound on the army leader (or just kill the army leader if he can't break the enemy), his minimum TPs awarded is 4 - and his maximum goes up to 6 if he can keep his opponent from doubling the number of models he can get off the board. For those who really like numbers, you can see in the graph below the count of distinct combinations of scores that have X number of VPs scored compared to the average TPs awarded for that score:

That graphic was very mathy - I apologize if I lost anyone in it. For anyone who likes math, you're welcome. To explain it simply, the blue columns show how many distinct scoring outcomes award either side X VPs (so the blue column for "5" has a blue scale value of 27 on the left because there are 27 distinct scoring combinations where 5 VPs are awarded - you can count this in the previous graphic if you like). The orange columns are the average number of Tournament Points you receive for the VPs you got (so the average score for anyone who got 5 VPs is about 11 TPs). 

My first takeaway is this: in general, you're basically getting the same number of TPs if you get 6-9 VPs (15-17 TPs, but this is a pretty steep jump up from the 11 TPs you get for only 5 VPs). Furthermore, most of the possible combinations will be in the 0-8 VPs scored, so there aren't that many distinct combinations that get you 9+ VPs (but if you get that high, you're getting between 16-19 TPs).

This is one of the less complicated scenarios, but as has already been mentioned, there are several scenarios in the Matched Play Guide have a similar scoring mechanic to this. If you extend this out to a few other edge cases (where you can get 1-3-5-7 VPs in one category), you actually include quite a few other scenarios in the mix. The most egregious scenario for distinct number of possible scoring combinations is Command the Battlefield, where there are the same 4 possible outcomes for breaking, the same 9 possible outcomes for wounding/kill the enemy army leader, and 5 possible outcomes for control of four different quadrants of the board. In total, this leads to 22,500 distinct combinations . . . I haven't bothered to compute how these look, by the way. I do have a life. :)

But this leads to a very important conclusion that I have yet to test, but I presume to be true: some scenario pools will be more interesting than others when using the German Scoring System. Let's walk through the different scenario pools and get a high-level view of what I'm talking about:

• Pool 1 has two scenarios with a limited number of scoring categories that have a lot of VPs awarded in an all-or-nothing way (Hold Ground has the same scoring potential as Reconnoitre and Heirlooms is not that different) and one scenario with a lot of scoring categories (Command the Battlefield, which as we mentioned has over twenty-two thousand distinct scoring outcomes).
  
• Pool 2 scenarios have lots of scoring categories that give 1-2 VPs per category, so you can get a wide spread of possible scoring results from it. This is, in my noob-level opinion, the best scenario pool for using the German Scoring System.
  
• Pool 3 is like Pool 1 - Destroy the Supplies gives you room for scoring variance (three "objectives" that are each worth 0 or 2 VPs, 0-2 VPs for having a banner, 0-2 VPs for wounding/killing the army leader, and 0-2 VPs for breaking the enemy), but Seize the Prize will be very all-or-nothing (scores similarly to Hold Ground and Reconnoitre). Retrieval will be either high variance or low variance depending on whether anyone can actually reach the flag and carry it off (it scores a lot like Assassination or Contest of Champions).
  
• Pool 4 is very all-or-nothing: To The Death has 7pts available for anyone who can quarter the enemy without being broken - all of which is off the table if no one breaks. Lords of Battle and Contest of Champions will give all-or-nothing to whichever army/leader kills more than the other guy with minimal points left over for the "losing side" (which is very similar to Recon or Hold Ground). This is, in my noob-level opinion, the worst scenario pool for using the German Scoring System -  but these games can be very one-sided no matter what the scoring system you use is, so that's not a knock against the German system in the slightest.
  
• Pool 5 has three very different scenarios: Reconnoitre has already been looked at, and Divide and Conquer is, in some ways, very similar - but it gives minimal points for breaking the enemy (0-1 VP without a dependency on your own force being broken) in order to give more weight to controlling the three objectives (3-5 VPs up for grabs for whoever holds the middle and 1-2 VPs up for grabs for whoever holds the two auxiliary objectives). This makes the game more dynamic than Recon (less all-or-nothing). Storm the Camp is notoriously hard not to get a draw (1-3 VPs for breaking, 1-3 VPs for wounding the army leader, but 0-3-6 VPs for holding the enemy camp AND holding your own - this gives us a total of 216 scoring combinations, or fewer than our "base case" of 252). I'm sorry to say this doesn't appear to get any better in the German Scoring System than in other systems - if you can pull a 6-0 win, you're getting the same major win you're getting in basically every other scoring system . . . and a draw gives both players the same TPs.
  
• Pool 6 has more variance, with Fog of War and Assassination often being independently scored (besides breaking), but can be less independent if the points level is low (since one or both armies might only have one hero, which narrows down who their assassin/protected hero is and who their target is). Clash by Moonlight is more all-or-nothing because of how killing more heroes works (though you get at least 1 VP for having a hero alive - so it can be better than Lords of Battle or To The Death).
  

I could probably work out what the distributions are for each of these, but I don't care enough to do that. :-) Let's wrap up this already-long post with some take-aways from our tournament.

Take-Aways from TMAT's 2023 Grand Tournament

I'm not going to go over all sixteen games here - we have a podcast episode that's being produced for that. Instead, I wanted to just provide a few anecdotal stories from the event to illustrate things we saw in the German Scoring System. First and foremost, we assigned a score of 14-6 to anyone who took a bye. I don't know what the usual assignment is for byes in European tournaments that use this system, but I calculated the weighted average score from the chart for all games that one side "won" and it came out to roughly a 14:6 win. I was like, "Okay, we'll go with that - you know, because we have no idea what an actual tournament score would be." A weighted distribution has the innate assumption that all scores are equally possible - and hopefully the previous discussion makes you think that's untrue, since some scores are just not realistically possible to get. However, we had to start from somewhere, so a weighted distribution seemed like a "fine" way to do it.

As it turned out, across the 16 games we played, the average tournament point scored by the winning player was roughly . . . 14 TPs. :) #NailedIt We also did a "bidding" system for the bye each round - each player could secretly vote whether they wanted the bye or not, but they could only get one bye during the tournament, and it went to the highest-ranked player that wanted the bye for the round (I always voted yes so that if everyone else wanted to play, they could . . . but as you'll see, this didn't matter much).

We also had a fool-proof tie-breaking system: ties would be broken based on who submitted their list to the tournament director (aka, me) first. Simple - no chances for tertiary ties and all the weight based on how easy you made the life of the guy running the tournament. This is the "Centaur special" for tie breaks - and it's the best system I know.

The scoring system saw the different players change positions quite a bit - to illustrate this, I'll give a brief walk-through of my games (note: this will include spoilers for the podcast - to avoid spoilers, jump to the "Conclusion"). I faced off against Rythbyrt on the first round and went from what could have been a 6-5 win for me in Capture and Control (I broke first, but had three of the objectives in my control) and slowly march down to a 4-8 loss as the game refused to end on me (BOOOO). 

This saw me at the bottom table against a guy I'd never played before (Sam) and after a very tight game of Divide and Conquer, I edged out a 4-3 win, with me having majority control of the center and one of the side objectives and Sam having sole control of the other objective. I only won because the game timed out - Sam would like me to emphasize that if the game had gone on for one more round, he would have won because a) his Gil-Galad was contesting the center and would have run down my Orcs, b) I was very close to breaking, and c) my Barrow-Wights were all out of Will. Sam's assessment is correct on all counts.

This minor win didn't bring me up very far, but my 1-1 record with close games put me above Sam and Centaur and I was in a three-way tie for fifth with Red Jacket and Tyler - more on this in a bit - and since I submitted my list first, I should have faced Strider, who was ranked #4 (since Dronak, one of the Top 3 players, took the bye that round). The problem was that Red Jacket and Tyler had just fought each other - they had the same score after Round 1 and pulled a draw against each other in Round 2. I decided not to enforce the strict rules of the tournament and make them fight each other AGAIN, and instead let one of them roll to see who got to face Strider and who got to face me. I ended up facing an army of crossbows and ballistas in Lords of Battle . . . you can guess how that went. 0-12 for me, thanks. :) Interestingly enough, the ballistas killed more models once I had guys within 6" of them than when I was 18" away from them . . . funny how shooting sideways works.

Since I got no TPs that round, I was solidly on the bottom of the pile and that meant that I got to face Centaur, who had lost 10-0 to Sam after being tabled (it was reported as a 12-0 because it was assumed that tabling meant a 12-0 was awarded, but I don't do that at my tournaments - I don't want to incentivize tabling because it isn't fun). We had a great game of Destroy the Supplies that was swinging between a draw, a win for me, or a win for Centaur until the very last fight of the very last round. Great fun was had, my Barrow-Wights finally showed up to play, and I lost at the last moment - but I didn't care at all.

As a spoiler for the podcast, I came in last at the tournament (which was better than not playing at all and foregoing the bye system), but there was quite a bit of movement all around. Rythbyrt surged to the top quickly and stayed there because he was able to keep winning, but Sam lost his first two games and ended up coming in third after two big wins in rounds 3 and 4. Red Jacket had a loss and a draw to start the tournament, but after one win and one bye, he finished second. The Black Prince (who I didn't face) won two games early, then lost two games later to finish in the middle of the pack. All in all, it was an interesting day and the German system made sure we didn't get a bunch of ties that needed to be broken in odd ways (though my tie was an interesting one).

The Impact of the German Scoring System - What-If Analysis 

Because I love math, I decided to compute what would have happened if we'd been seeded by the German system, but given scores in our normal system (2-to-10 TPs, 6+ VPs required for a major win, 7 VPs for a bye) and there was some movement within the middle of the pack. Some player positions wouldn't have changed - Rythbyrt would have still been the undisputed winner, I would have still been the undisputed loser, and Tyler would still have ended in the direct middle - but the order for #2 through #4 would have changed (Red Jacket would have ended third instead of second, Dronak would have ended second instead of fourth, and Sam would have ended fourth instead of third - and this was all because of how the points were awarded when fights were borderline between major and minor wins). Similarly, the guys in positions #6 through #8 would have changed a bit just because of how the scoring worked. Still, it's not like someone jumped from the bottom to the top just because of the scoring system used - the rankings were still consistent (it was just how you placed relative to people who did about the same as you). And I think the greatest benefit of this system is found in its ability to provide granular value to the VPs you're getting (or your opponent is not getting) in each game.

Conclusion

Hopefully you enjoyed this post - and if you run tournaments but have never used this system, I would highly encourage you to try it! If you've participated in tournaments that use this system, I'd love to hear your thoughts on it! Until next time, happy hobbying!