France 1944 Fixes

France 1944 box cover France 1944 is a conflict simulation published by Victory Games in 1986. The simulation covers the Allied advance on Germany after the D-Day landings in Normandy. It is a notable simulation, not for its topic, but for the innovative (at the time) mechanisms it introduced for simulating simultaneity of movement and combat in a two-player game as well as the limitations of command and control. Nonetheless, many players have found the game's combat resolution system unsatisfying in some specific instances, even while enjoying and praising the game's many effective design concepts.

As the designer explains, the original combat system attempted to show how combat efficiency (in the form of training, experience, and morale) can have a greater impact than numerical odds, as often happened in the battle for the bocage, the subsequent breakout, and the toughened defense of Germany. The design also incorporated coordination issues into the system by assuming that higher force ratios reflected more complicated operations. Although the simulation goals were completely sound, the combat resolution system did not resonate for some players in the specific case where equal quality (in game terms, morale) units were encouraged to attack each other at 1:4 odds to get the best possible result while ensuring the attacker remained unscathed (an 8 in 9 chance). It was my original opinion that perceived problems with the system resulted from linking coordination effects to force ratio instead of absolute force levels.[1] Furthermore, the system treats combat efficiency (i.e., morale) effects as a bimodal function instead of as a function of combat efficiency differential. Note, this is not the same as complaining that the CRT should be linear or that the entire CRT is broken—such sweeping generalizations do not reflect my opinion. An alternative way to look at the issue is that the boundary line of the CRT—in fact, it's actually the Combat Result Determination Table (CRDT), not the CRT, that is at issue—between superior and inferior performance is too generous for equal quality unit combats. In effect, the CRDT needs to be bloodier.

As a specific example, consider the 1:4 column of the CRT, where the vast center of the Gaussian distribution produces a C result on the Combat Result Determination Table for a 3 morale attacker against a 3 morale defender (a 32 in 36 or 8 in 9 chance). On a C combat result, the defender loses 1 step (and retreats) or 2 steps (and stands fast) while the attacker remains unscathed (yes, at 1:4 odds—or even 1:20 odds). After such a 1:4 attack, the attacker may now find itself attacking the exact same defender at a higher odds level where D results become increasingly likely. A D result has the same effect as a C result except the attacker suffers 1 step loss. Assuming the model is correct and a 1:4 attack is far more likely to do a lot of harm to the attacker while leaving the defender unharmed, then it is unrealistic that a commander would attack at greater than 1:4 odds (in the 3 morale versus 3 morale example) just because that's what the force levels on the counters mandate. The commander would allocate only a fraction of his available forces to attack. Therefore, the player should at least have the option of attacking at lower odds than the calculated combat ratio. Even if it were possible to break a unit down into its constituent subunits to make a lower odds attack, it would be tedious for the player to engage in such factor counting when a simple rule modification would have the same effect.[2]

Over time, I developed some optional rules for France 1944 that attempt to stay faithful to the game's original design goals. Although primarily focused on the combat resolution system, the changes also address a potential play balance issue with respect to reaction point expenditures.

Instead of using the revised CRT, you may want to try retaining the original CRT while prohibiting attacks at less than 1:4 odds and allowing an attacker to use any odds column for a ratio less than or equal to the computed odds ratio instead of mandating that the column for the computed odds ratio be used. In addition—or as an alternative—you may change the C combat result to cause the attacker to lose 1 step when the defender loses 2 steps.

The original CRDT is not bloody enough for the attacker when the attacker and defender share the same morale, causing a narrative discontinuity when an attacking unit becomes more likely to lose a step when forced to attack at higher odds immediately after making a low odds attack against the exact same newly-reduced defender. Given that these anomalies occur on a particular boundary within the CRDT, a satisfactory solution may be achieved by using the result columns for the ratio result one greater (e.g., use the result columns for 2 to determine the results for 1, but retain the defender morale reference column for 1). That requires adjusting the results for ratio result 5 following the pattern used by the rest of the table. For example, the ratio result 5 set-piece combat rows become D D D D D C and A A A A A B.

I will digress a moment to point out that upon the publication of Washington's War, its design notes revealed that the battle resolution system in We the People was flawed in the sense that it did not produce the results that the designer had intended. Despite that flaw, many players enjoyed We the People. I was not one of them. In an article appearing in C3i Nr24, the designer admits … there were many things that bothered me about some aspects of its historical simulation that kept me from playing it often.[3] I felt the same. Thankfully, the designer had the opportunity to redesign the game and produce Washington's War, which I think is a great game that fixes everything I found lacking in We the People. My changes to France 1944 in 2003 were nothing more than an attempt to turn an existing game that didn't work for me as well as it could have into a game I want to play over and over again, the same way that the designer later turned We the People into a game he wanted to play repeatedly by creating Washington's War. Just as people who enjoyed playing We the People repeatedly weren't bothered that the designer didn't share their enthusiasm for the game, players who enjoy playing France 1944 without any changes shouldn't be bothered that others don't share their enthusiasm.

France 1944 Revisionshtmlpdf

Before evaluating these modifications, you should play the game without any changes. If you like it as is, great! If you think the game requires improvements, then try out these modifications. If you do not have any issues with the original combat resolution system, keep it and use the other less-controversial modifications.

A redesigned version of the game was released in 2020. The designer responded to a customer's question about the new combat system as follows:


As this is the internet I no longer debate stuff like this as it takes too much effort and it changes no minds. Witness the original debate where the more is better is the basis of the argument.

 --France 1944 designer responding to customer on ConsimWorld

I will note that the original debate did not have more is better as the basis of the argument. If the designer had been more willing to listen to players (and actually played the game as much as his customers) instead of putting words in their mouths, it wouldn't have taken him decades to understand what needed improvement. Based on a reading of the online rules and an examinaition of the combat matrix, it is clear that the designer fixed the problem of low odds attacks being relatively risk-free for the attacker. Overall, the new combat system is bloodier, fixing the issue some players had with the original game. Unfortunately, the casting of aspersions on those players has not ceased. Although the new combat system may seem more convoluted to some, the combat matrix makes it clear what the design is doing. The success or failure of an attack is a bimodal function of morale—essentially a coin toss at 3 morale with no modifiers—but the defender also must determine if his defense succeeds or fails. Within the range of results, attacks at too-low or too-high odds result in more attacker step losses, fixing the problem with the original game where too-low odds attacks resulted in no attacker step losses. In addition, higher odds attacks result in additional defender step losses. In the face of a failed defense, an attacker may even cancel out a step loss at higher odds.

Instead of using the revisions provided above, you may want to try retrofitting the new game's combat system into the original game (or simply buy the new game after reading the new rules). But you will need a scan of the new combat matrix and the following table in the absence of the custom hit die:

Table 1. Hit Table
Die RollHits

Wargame design and development is imperfect. Designers and developers are human and can make mistakes. The designer in question had previously been partially responsible for another problematic CRT in a game he developed for none other than designer James F. Dunnigan (who—despite his legendary status—was primarily responsible for the problematic CRT).


The CRT was a serious design mistake, which should have been caught earlier, according to [the developer]. I questioned the table much earlier than the playtesters did, but when I did a few quick calculations, it worked, because I always started in the middle of the table, and there it does work. Only later did I realize it didn't work at the ends of the table.

 --Design and Development: The Next War as Supergame”. Stephen Donaldson. Moves 42, (December/January 1979), p. 11.

Sound familiar? This is the way wargame development actually works. Don't assume everything in the box is perfect and don't begrudge someone for trying to make it better. After paying an arm and a leg for a wargame, you've earned the right to tinker with the design.

[1] This analysis has nothing to do with an expectation of a naïvely linear combat results table. I completely understand the designer's explanation of the reasoning behind the design of the combat results table. More importantly, I understand and agree completely with what it is supposed to do; I merely think it needs a small tweak to achieve its purpose.

The crux of the issue is found in the following response the designer made to a player on ConsimWorld: Yes, the poorer unit is doing better in a less complex situation, so it makes the crt look wrong. I believe that in the aggregate it captures the historical dynamic, but if you don't think so, so be it. The design attempts to produce a globally correct result via an accumulation of seemingly incorrect local results. It is those local anomalies that destroy the overall narrative of the game for some players.

Compounding some players' confusion, the game's player notes recommend the Allies make high-odds attacks and the rules themselves require to round odds in favor of the defender, giving the impression that higher odds are more favorable to the attacker. Yet, there are clearly instances where rounding odds in favor of the defender would actually require you to round odds up instead of down. If the rules had been edited with care to present a vision consistent with the designer's intent, and if the rules had contained proper design notes (not the blurb on the back of the box) explaining the design instead of player notes that appear to be inconsistent with the designer's stated intent, fewer players would have found the game's combat results confusing.

[2] The designer's response to this observation was, I cannot help the fact that you can see the combinations. But the CRT is forcing the right decisions.

At the scale of the game, higher odds need not necessarily correspond to greater attacker density. Just because a counter has a particular strength factor doesn't mean the commander of the forces represented by that counter would commit all of the forces in an attack. The obvious solution to this problem is to allow an attacker to use any odds column up to and including the computed odds ratio instead of mandating that the column for the computed odds ratio be used. If the attacker has 3:1 odds against a defender, he should be allowed to use the 2:1 or 1:1 odds columns instead of being forced to use the 3:1 column (often the 1:4 column is the most advantageous). This mitigates the problem while retaining the original CRT, but it does not resolve the underlying problem that you have to deconstruct the behavior of the CRT and let the CRT drive your decisions instead of focusing on the operational strategy without fixating on the idiosyncracies of the CRT. This problem could be said to be a feature of all wargames using CRTs, but ideally it doesn't rise to the level of reducing a game to an exercise in factor counting.

[3] It is therefore perplexing that in another piece in the same magazine issue, he denigrates players who refer to aspects of wargames as ahistorical on the Internet, when he has made it clear he feels aspects of We the People are ahistorical (though he is careful not to use that particular word and not do it on the Internet). Lest this observation be misconstrued, I want to make it clear that I am a fan of the designer's games and am grateful for all of his contributions to the wargaming hobby. I simply don't understand the inconsistency of the designer's comments or his repeated denigration of anyone who dares critique the probability distribution of combat results produced by the game.