Original Design by Chris Perello
Reviewed by Peter Perla,
interrupted by random musings from the editor
You guys all know what a paradigm is, don’t you? (Granted, some of you don’t even know what a pair of socks is…at least not a clean one). It’s a fundamental way of looking at things; a model of reality. For the past twenty years, the “Great Battles of the American Civil War” system, in its many, and varied, incarnations, has been the paradigm of regimental-level Civil War battle games. Chris Perello and XTR, in a much ballyhooed break from that tradition, propose a new paradigm in Fateful Lightning (or as we lovingly call it, FatLite, a delightfully oxymoronic appellation if there ever was one). The game is quirky, irritating and incomplete, but once you figure it out, you discover an amazingly subtle system that works very well. I had some strong initial misgivings, but after A LOT of work, I now believe that Chris may, indeed, have latched onto something here. Unfortunately, the game has an unfinished feeling about it, one that XTR would do well to dispel with a Second Edition rules book and charts. Nearly everything about the game is different from what went before, sometimes obviously, sometimes less so. The game is played in Turns representing about 3 hours each during the daylight. Each Turn is composed of one or more Couplets. Couplets are where the action takes place, but with a twist. At the start of each couplet, the players roll a die to determine whether the Player who moves first remains the same or changes. If there is a change, the player who would become the new first player may instead choose to end the Turn. In addition, on certain die rolls, the turn ends automatically. Thus, a Turn may last from no couplets to an unlimited number of couplets. This adds a bit of uncertainty and extra spice to the situation. [Ed.…and sounds a mite like the TCT system, a mechanic one would think XTR would abjure like the plague.] Other basic precepts are also modified or changed dramatically. The regiments and batteries are not rated for strength points and do not suffer the types of incremental losses so familiar in TSS. Instead, most infantry regiments have the same effective strength (2). Units can stack up to three deep, but only the top infantry unit and one artillery battery may be involved in normal combat. There are actually three flavors of combat: “regular” combat, representing the stand-up firefight at 100 to 200 yards; bombardment combat, representing long-range artillery and small arms fire (most rifle-armed units have a range of 2 hexes, or about 400 yards); and charge combat, representing — well, charges, of both infantry and cavalry. Perhaps the most obvious and most significant difference between FatLite and previous approaches to this level of combat lies in the combat resolution system. Regular combat and bombardment use the same table, based on a combat differential. Units may be disrupted, routed, captured, or eliminated (including exchanges). If you are in a hurry to take ground and don’t care what it might cost, you can opt for the charge. A charge is almost like a blitzkrieg attack. First a morale check determines whether the charge is successful. If not, the attacker usually takes it in the ear and the fight is over. If the charge succeeds, watch out! Each attacking unit attacks each defending unit in turn until the attacker is disrupted or eliminated or the defender’s hex is emptied. In the latter case, the charging stack may continue to attack any adjacent enemy stacks, automatically using the successful charge table. Did you ever want to roll up an enemy line? This game allows you to do it in fine style. It also allows you to destroy your army in a heartbeat by going charge-happy. Actual effects of combat usually depend on the unit’s morale rating. The intriguing thing about the table is that there are two possible outcomes for each die roll. If the first result has no effect, then the second applies. For example, with a +2 differential, a die roll of 3 yields the result A6(Ex). If the attacking unit’s morale is less than 6, it becomes disrupted. If all attacking units have morale of 6 or more, then the Ex result applies. The side with fewer units involved loses its units, and the other side must remove an equal number. But an “eliminated” unit is not gone forever. In what is probably the most controversial element of the system, Perello has included a mechanism for resurrecting such units (Magic Player Alert!). This mechanism revolves around a concept called Extra Strength (ES). Most formations in the game (usually a formation is a brigade) have some number of ES points available to them. By expending an ES point, a player may return an eliminated unit to play during any of his subsequent movement phases. Such units must be placed adjacent to a friendly unit in their formation (but not adjacent to an enemy unit), the friendly leader in whose command they fall (if one exists), or the friendly “train,” (the army’s supply wagons). Chris describes his reasoning in a Design Note: “Elimination does not mean every man in the unit has been killed, but that enough men and leaders have been hit to render the unit ineffective.” He then explains that “Each counter represents only the front line of a unit. Any men beyond that are represented by the extra strength markers.” My first (and second) reaction to this was negative. “Smacks too much of ‘design for effect,’” methinks. Surprisingly, the more experience I had with the game, the more effective and realistic this approach seemed. It can still be abused — if you have a formation widely split, for example, a unit can be eliminated on one flank of the army and reappear immediately on the other flank, miles away. But generally that doesn’t seem to happen because you usually want to bring the unit back close to where it was killed. [Editorial Epiphany #1: I had reservations about this particular mechanic, mostly because a unit so rendered ineffective did not simply disappear, nor was it easy - or even desirable - to extract such a unit. Using this system theory, the 20th Maine, having taken a certain number of losses, would have melted away … something which we know it did not do. That the mechanic works “in the end” is simply another application of design for Effect.] Another thing that I and my fellow players scoffed at was the lack of detailed command and control rules. There are some commanders present, but not for every formation. (For example, there is no Pickett because in the designer’s estimation he did little but pass on orders.) Gone are the familiar notions of command range. Again, however, first impressions were misleading. The commanders that are present are extremely important. They can use their command ratings to carry out all sorts of important functions, from urging their troops forward (adding to the base movement allowance), to rallying the routed, to forming artillery Grand batteries to pulverize the enemy line. They are also important in light of the Army Initiative rules. In fact, the one rule that specifically reflects command and control is the Army Initiative rule. Somewhat reminiscent of a similar rule in Perello’s Lee’s Greatest Gamble, this rule is a key component of the game’s feel. There are three possible initiative states: Full, Partial, and Low. Under Full initiative, all leaders are Active (that is, they may use their command capabilities), and combat units have four movement points and are free to enter enemy ZOCs. Under Partial initiative, the leaders are Active and units may enter enemy ZOCs, but the base movement allowance of units is only one movement point. (Here is where the ability of commanders to add movement points to friendly units under their command becomes crucial. Many a promising attack can stop dead in its tracks if the initiative drops to partial and there is no friendly leader available to spur his men on.) Finally, Low Initiative really puts the damper on things. Units have a movement allowance of one, they may not enter enemy infantry ZOCs except for special circumstances, and all leaders are inactive until they roll a wake up call against their activation rating. Low initiative prevents starting an effective attack, but does not necessarily kill one that’s under way, especially if you can wake up the key commanders. A most interesting aspect of the Army Initiative rules is that the two sides have different characteristics. The Confederates have a good chance at Full initiative and about the same chance of Low initiative. The Union, on the other hand, has low probabilities for either Full or Low initiative, and much greater chances for Partial. Thus, players must adapt their plans to the characteristics of their army. I’m not sure that I completely agree with this “all-in-one” approach, though. Perhaps applying the Initiative states to individual corps might not better represent the events of 1863. [Editor’s Epiphany #2: “Hear, hear!” Corps level Initiative ranged widely throughout this battle. and was one of the problems with which the Overall Commanders had to deal. Meade, for the most part, let his individual corps commanders run their own show.] But I must admit that the overall effect of this rule is amazingly historical in its feel. Thus, despite the lack of detailed command structures and traditional command rules, and with the addition of some command-based stacking and coordination rules, the game seems to succeed in requiring players to treat their formations as formations, not merely collections of independent counters. In the dynamics of play, you find yourself maintaining your command structures, not because of some artificial rules that force you to do so, but because doing so makes things easier for you, the player, to control your cardboard army and use it effectively. If this is Design for Effect, then I’m here to tell you that the effect is to make you think and act like Civil War army and corps commanders. It ain’t perfect, but it sure do work. There are lots more new ideas in FatLite, some of which were previewed in Lee’s Greatest Gamble. One of the more interesting is a nice variation on random events called the Fortunes of War rule. Every time you roll a ‘0’ in resolving combat, one player gets a “Fortune” to spend immediately. You can use your Fortunes to take a pot shot at an enemy leader, move an enemy unit (a perennial Perello ploy), rally a routed unit, and more. There are sharpshooters, cavalry delay tactics, and field works. There are even “heroes,” who provide charge and combat morale bonuses but tend to die fairly quickly - a nice historical touch. I think my experience with playing the game is probably not atypical of gamers weaned on TSS. I got together with two, of my fellow Perlas Before Swine Playtesters, Gregg Adams and Ed McGrady. We played the opening hours of the battle…and we hated it. The Confederate infantry had a movement allowance of 4, while Buford’s cavalry only had an allowance of 2 (the Confederates naturally had Full Initiative and the Union only Partial — a fairly typical - [Ed. but, for this particular point in the battle, rather ahistorical] situation). During our first playthrough, Buford deployed across Herr ridge just as he would in TSS. Heth attacked and routed virtually every unit in the line. I certainly did not like this result, thinking it entirely unrealistic. “Pshaw,” we shouted. “This is not the Civil War, this is more like Pennsylvania panzers.” I got on GEnie and attacked the game, at the same time posting a list of questions to Chris. I was not happy. Then, something happened. As Lincoln wrote to Grant in 1864, “I begin to see it.” After studying the rules and the map more carefully (and it took no small amount of time, either), I realized that the problem was in how I perceived Buford’s delaying action should be represented on the board. In FatLite, Buford delays Heth by preventing the Confederates from using the triple-speed road movement rate (available only if a unit starts and ends its move on the road, moving only through road hexes, and never adjacent to an enemy ZOC). But Buford’s boys must also stay out of reach of the Confederate infantry for as long as possible. They cannot stand up to a full scale assault for long, and there is no Extras Strength for them to rebuild. Only by careful disposition and maneuvering, and by standing and fighting as little as possible, can Buford do the trick. With good play, however, it is possible to hold on to MacPherson’s Ridge until Reynolds and the I corps arrive. [Editor’s Epiphany #3: Design for Effect, again. The delaying of Heth by such a mechanic may be the end result, but it does not reflect the Reason Why Heth’s initial attack was so desultory. Heth was (a) operating under somewhat vague orders from Lee; (b) was not a great - or even good - division commander; and, ultimately, (c) committed units piecemeal until it was too late. The “reason’ for Heth’s delay then, was, at least in great part, do to his command inabilities. The FatLite system does not cover this aspect.] To get them there in time, I figured out how the Reserve rules and the leadership movement bonus rules allow Reynolds to recreate the feat of bringing Meredith and Cutler into the battle in the nick of time. (I will leave that as an exercise for the reader.) Suffice it to say, that, after a foolishly overly aggressive Confederate move that coincided with the arrival of the Black Hat, I lost all of Davis’ brigade in a single couplet. I also tried out Pickett’s charge, using my own setup because that provided in the game was from the planet Pluto (subsequent errata has fixed most of the problems). With the entire Union II Corps massed along the ridge and supported by artillery, the chances for success seem appropriately nonexistent for the gallant rebels. The latter broke themselves against the bayonets and canister fire of the Union line. I consistently wrecked the front line stacks of three Confederate regiments with relatively few Union casualties and many opportunities for counter charges against disrupted attackers. But I got little of the sense of the ordeal of the Confederate infantry trying to charge under Union fire the whole way across the fields in front of the ridge. FatLite seems to work OK at the big picture level, but Pickett’s Charge also highlighted the awkwardness of the interaction of map and unit scale. That interaction works well for the combat model, but at a high cost in clumsiness of play. There were a LOT of three- or four-high stacks crammed adjacent to each other. Essentially you are playing with as many units as in TSS, but on a battle area roughly half as large. It makes for many heavily stacked hexes and almost demands you master that set of tweezers languishing in your back drawer somewhere. However, a similar experiment with Hood’s Day Two attack proved even more surprising and illuminating. The combat system and the use of Extra Strength combined to produce an extremely faithful recreation of the battle for Devils Den and the Wheatfield. Having learned my lesson about rash charges, I used a more methodical Confederate attack patterned along the lines of the actual battle. After some initial difficulties, the Confederate infantry of Law’s and Robertson broke the Union left in the Devils Den area. The Union had to rebuild its units farther to the north and east, successfully simulating the retreat of Ward’s brigade up the Plum Run Valley. I was impressed. Finally, Gregg Adams and I took another crack at day one. Gregg used the same tactics as in our first go round, charging the forward cavalry units once he closed the range. But instead of delaying, I had those units stand and fight, because I had realized that a charge was more likely to kill his units than mine. Lo and behold, Davis’ brigade was wrecked (three units disrupted, one routed, and two eliminated, and Davis with no ES to bring them back). I was treated to a sight I had never seen in any Gettysburg game I had played before: the Confederates retreating back toward their entry hex. Bad tactics and rash charges had been duly punished. There are a lot of things Chris Perello has done in FatLite that I would still question. There are more holes in the rules than in a tree on a Civil War battlefield. Yet, once you free your mind of its preconceptions, you can begin to understand what the rules are trying to say. And you realize that there are a lot of interesting design ideas here. The worst part of it is that the development effort is in the best traditions of 3W, not XTR. This results in scenario setups that are limited to the beginnings of each day only, preventing you from easily exploring Longstreet’s assault and Pickett’s charge. (In fact, the scenario for the third day’s fighting is completely screwed up and you will have to get the errata from XTR before playing it.) The game maps cover not only the usual battlefield, but extend east as far as the cavalry battlefield near Hanover. Unfortunately, the map stretches the envelope for ugly. There are some very interesting -- and subtle -- rules about the creation and use of reserves. And you will look on combat leaders and their uses in entirely new ways. Most of all, you will be forced to think about Civil War combat and its representation on the game board in ways that TSS did not reflect. We have all had a vague concern that casualties in our games always seem much higher than they were in the real world. We also brush that off by attributing it to how we play, rather than to the combat model itself. FatLite calls into question the TSS model and all subsequent strength-point-based systems. It’s about time. Let the debate begin anew. We still have much to learn. Editor’s Epilogue: OK, so why the title? Peter, being a nice guy, has shown “admirable restraint” in not mentioning it, but neither I, nor BROG, can ever qualify under those two words. FatLite contains one of the classic “boo-boos” of gaming graphic history, one that only succeeds in adding mystique to what we, here, considered one of the least attractive maps in decades: they left the “B” out of the “Gettysburg” town name. This is, to be sure, an extremely trivial matter in what is an important and rather clever design. Does it affect play? Not in the least… but it sure is worth yukking about. And my apologies to my good friend, Beth Queman, for whom I promised much “Tea and Sympathy”. CAPSULE COMMENTSGraphical presentation: Okay, but nothing to write home about. The counters are quite nice, but, as one wag on GEnie pointed out, some of them look they are wearing pajamas. Playability: This is not an easy game to learn or master. Vagaries, abundant errata, and rules holes don’t help. Tweezers essential, and it is long, especially for new players. Replayability: A strong point. The couplet system creates tension and uncertainty and can make for vastly different games. Wristage: Lots, but not as much as the GBACW games. Creativity: This ain’t your father’s TSS. Lots that’s new and different here. You will have to rethink what you thought you knew about warfare at this level. Historicity: Far better in overall feel than in individual details. Comparison: A vastly different model and feel for tactical combat than TSS. To this reviewer, a more realistic-feeling model for the Civil War. The Gamers’ Thunder at the Crossroads covers roughly the same geographic area with brigades rather than regiments. Once you master the rules (no mean feat) this is easier and faster to play than both of the others. Overall: Important but flawed. If you want a quick and easy game at this scale, this is not it. But if you have a serious interest in Civil War battle tactics in general, and Gettysburg in particular, then you should definitely experience Fateful Lightning. Here’s hoping Ty comes out with a badly needed second edition. from XTR Corporation
Back to Berg's Review of Games Vol. II # 16 Table of Contents Back to Berg's Review of Games List of Issues Back to MagWeb Master Magazine List © Copyright 1993 by Richard Berg This article appears in MagWeb.com (Magazine Web) on the Internet World Wide Web. Other articles from military history and related magazines are available at http://www.magweb.com |