Analysis of a TPK

OK, technically it wasn’t a true TPK since one of the characters got away. So I think I can still say with a straight face that I’ve never had a TPK. But to the extent that it wasted a ton of the players’ character development efforts, and a lot of my campaign planning, it was pretty much a TPK. A campaign reset. This is kind of a big deal – it’s not fun for anybody, it wastes a lot of work, and may be reflective of a problem. It’s especially alarming since another character was killed a couple sessions previously (to be fair, that one was just a random crit from a BBEG for a truck-load of damage). So I thought I’d take a little bit of time to understand it.

The background:
The group consists of a bunch of guys who haven’t played an RPG in 1 or more decades, much less 4e D&D. Mortgages, wives, babies, careers, graying hair, the works (I’m the odd guy out – a regular gamer for years, no mortgage, no baby, no gray, and I don’t think you can call freelancing a career. Thank god I’m married though.). We sit around, grill, laugh, drink and game. Their characters are all level 2 and consist of a Fighter, Cleric, dual-wield Ranger, Wizard, and Warlock.

I’m running the players through the Keep on the Shadowfell module and the players get to the kobold lair (spoiler alert, don’t read on if you are planning to play). In the module, the party runs into a large cluster of Kobolds outside the lair, and presumably one of them escapes into the lair to sound the alarm. The party sends its stealthier members forward to spring an ambush on the unsuspecting kobolds, but unfortunately this goes awry as Tostig the Fighter slips on loose rocks near the creek-bed. The party sprigs into action and begins to slaughter the poor kobolds.

I think I play the kobolds as being pretty smart and cunning. They fall back and take cover leaving the PC’s long-range attacks without good targets. They dart out, hurl javelins, and then shift back into cover. Some kobolds start sneaking around the flanks. When the PCs get strung out of formation, they rush forward and mob a single character to maximize their mob attack bonuses, combat advantage bonuses, sneak attacks, and concentrated damage. But ultimately, the characters are Heroes and the kobolds are, well... kobolds. They get mowed down. But the Kobold Slinger gets away to raise the alarm. The party splits up to gather thrown weapons, search for loot, etc.

Red Flag #1: when the kobolds mobbed Tostig, they did a large amount of damage – managing to bloody him. This should have alerted both the players and myself that “Hey! Being surrounded by kobolds just might get you killed!” But other than that one incident, quickly rectified, the kobolds weren’t really a threat.

Red Flag #2: drinking. I definitely enjoy having beers while we game and so do the players. But I do wonder if that clouds our judgment too much and we miss important cues like the above.

Red Flag #3: metagaming. Everyone has now played about a half-dozen sessions of 4e D&D and has come to expect the post-battle recharge – regain all your encounter powers, heal HP, etc. They feel comfortable with splitting up and letting their guard down. What the players don’t know is that when the Slinger ran away, that started a countdown to reinforcements. They aren’t going to get a complete rest and they squandered some encounter powers in the first part of the battle.

A couple rounds later, the first wave of reinforcements comes streaming out of the cave. OK, those of you who have read the module know that I’ve deviated a little from the script. Mid-way through the battle I went back and reread the encounter description. Somehow I ended up with the impression that if one of the kobolds got away to raise the alarm, the kobold reinforcements came outside in two waves. Otherwise, the two waves were encountered inside. Oops. In the module the party is supposed to be able to rest after taking care of the kobolds outside. However, because the party was still kicking ass and taking names, albeit strange hissing- and yipping-sounding kobold names, I didn’t worry about it too much. And the party was already level 2 due to some earlier adventures and the module is intended for level 1 characters. In my head, I decided to weaken Irontooth a bit (those 100+ HP BBEG just get boring to fight after a while anyway) to balance it out.

DM Mistake #1: the first real outright mistake was not giving the PCs a rest between waves. Even with the party being level 2, they just didn’t have enough powers to wade through that many foes.

So, the first kobold wave comes streaming out of the cave. They split up, half engage the PCs directly, the other half move off a ways to hurl javelins at exposed targets. The party is caught out of formation and the kobolds gun for the “squishies” like the Wizard and the Warlock. Thankfully the players are able to get out OK and get the Fighter, Ranger, and Cleric up front. A back-and-forth melee ensues where the players continually cut down the kobolds. The players were continually mobbed by the concentrated attacks of the kobolds, and took a lot of damage, but nobody went down. So the second wave started off a little sketchy but the players rebounded.

When there were a handful of the wave two kobolds left, wave three arrived. At this point, the players were all still above bloodied and handling the encounter fine so I still wasn’t worried (in fact, I was vocally rooting for the kobolds because they were getting annihilated). But the situation went bad really, really quick. The party still hadn’t found a way to restrict the kobold movement. When possible, I had them avoid the axe-wielding armored horror that was Tostig the Fighter, and instead attack the weaker party members. The Wizard strayed too close and got sacked by a mob of kobolds – down he went. The Warlock was mobbed and dropped right after. A couple attempts to heal the fallen members were considered, but the party came to the conclusion that they’d just get immediately killed again, so the healing was reserved for those who were still standing. It was a moot point since the kobolds were finally able to completely surround Tostig and with the massive mob bonuses and concentrated damage, whittle him down. That left the Cleric and the Ranger who attempted to run. The heavily armored Cleric didn’t make it. Surprisingly, Irontooth took so long to wade into the fight that he had almost no impact – he only landed one solid hit.

DM Mistake #2: I don’t yet know 4e well enough to know when the players are in trouble before it’s too late. It’s easy to know the party is in trouble when half of them are bloodied or dead. A good DM knows this is going to happen several rounds in advance, and can take action accordingly (even if it’s only to remind the players that they can attempt to run away).

PC Mistake #1: never let cunning foes get at your squishies. They know this from nearly having lost the Wizard once before. There are a couple chokepoints on the map that the players never tried to occupy that could have limited kobold movement.

PC Mistake #2: party composition. I got the impression that the party had very little battlefield control. There was only one Defender, so the front line wasn’t very “sticky”. Both the Wizard and the Cleric focused more on direct damage powers than they did on control or healing powers. And only one of the characters was a truly melee character, with decent AC and HP. The Wizard and Warlock were of course ranged attackers, the Cleric was OK in melee but had better ranged powers, and while the Ranger had the AC and HP to support a more “front-line” role he preferred hit and run tactics. So this created two major issues: the kobolds could get at the more vulnerable party members fairly easily, and the single melee character was simply overwhelmed without support.

PC Mistake #3: maximizing short-term damage over concentrated fire. This may not be an entirely accurate recollection, but I swear I saw the party members repeatedly choose to make attacks that dealt the most damage to different foes, rather than concentrate fire on a single target till they drop. For example, I believe the Ranger and the Warlock repeatedly attacked targets that were closest to them so they could get the extra damage, rather than attacking the more distant kobold that Tostig had hit. Likewise, on a significant round the Wizard opted to do an AOE attack on a cluster of kobolds, which certainly was a nice amount of damage, it would have been better to pick off some of the wounded ones already engaged in melee. On any given round, the 5 players would seemingly attack 3 different targets.


In summary, I think I made a big mistake, but it was exacerbated by the fact that the players are still learning how to play 4e. They don’t yet have good tactics and party synergy. Interestingly, none of us had much warning that the encounter was going to go so bad – it happened very suddenly. The party went from slaughtering the kobolds to getting slaughtered in the course of two rounds. I think it was a combination of decent rolls and the party running out of steam (encounter/daily powers). Because it happened so fast, the party didn’t even know they were in trouble until it was too late.

I hope you find this review useful. Take the time to review your games, especially when they go wrong. I learned a lot by examining this. As a DM or a player, you will make mistakes. Learn from them.

SE1: first attempt is a failure!

Huh. You know how I said the timing of the Oracle in City 1 and the completion of the Forge in City 2 is crucial? Yeah, it's a pain in the keshik. Let's review the final steps in the gambit:

1. Acquire Priesthood
2. City 1 whips a building to completion that had been previously halted with 1 turn of construction remaining
3. City 1 begins building Oracle, benefiting from the overflow of step 2
4. City 1 whips/chops to finish Oracle quickly
5. Choose Metal Casting as the free tech
6. City 2 whips another building with 1 turn remaining
7. City 2 begins building Forge
8. City 2 whips/chops to finish Forge WITHIN 6 TURNS OF COMPLETING THE ORACLE (this is why pre-chopping the forests is so important)
9. City 2 tweaks its population so that it has an Engineer Specialist
10. City 2 will eventually spawn a Great Engineer
11. City 2 begins building Pyramids
12. 'Pop' the Great Engineer to build the Pyramids
13. Convert to Representation, Caste System, drop your Research Slider to 0%, maximize city populations (without starvation/unhappiness) and use tons of Scientist Specialists

Congratulations. If you did it right, you should know be cranking out tons of research, and making good money to boot. Great Scientists will be popping out like crazy. And you can do this all by 1000 BC!

Or you can be like me and pull it off by 325 BC... Not so good then. Even though I got Priesthood and access to the Oracle in 2000 BC, I could not get the Forge built in my second city for quite some time. I hadn't been growing the population fast enough in City 2 to allow for two whips in rapid succession and still have population left over for an Engineer Specialist. I also needed a mine in my second city - once those forests were chopped down, there weren't a whole lot of hammers around.

A little late?

The other issue I encountered is the population of my first city would occasionally increase while I was waiting for the Great Engineer to spawn. This new pop would invariably be a Scientist and if I didn't catch it right away, it would destroy the careful Great People Point (GPP) planning. I would end up with a Great Prophet spawned in my first city (from all the GPPs generated by the Oracle) rather than the Great Engineer. This is also the same reason why you must finish the Forge and assign an Engineer within 6 turns of completing the Oracle - if you take even one more turn, your first city will generate enough GPP to create a Great Prophet first.

So, I must have reloaded from 2080 BC and made tweaks 10 times. The best I could get is the Pyramid at 325 BC. I suppose I could have reloaded earlier to around 2700 BC and done a better job but I got rather frustrated and would rather start over.

This was a fun experiment, and is highly recommended. It was very cool to move the Research Slider to 0% and still crank out research points like crazy.

SE1: to 2000BC

Wow, the past millennium just flew by. Not a ton happened. I continued to explore but alas, only one more goodie hut (48 gold). Peter is indeed waaay to the north, so I can ignore him for a while. No religions discovered in my hood apparently.



I built Bombay on my chosen spot and built a Fast Worker. My two workers are rapidly roading and pre-chopping according to plan. IMPORTANT NOTE: I misread the worker's estimated time to completion. When it says 3 rounds to chop down a forest, that means [The Current Round] + [Next Turn] = Finished. So to do this right, cancel the order the very next turn. I had to reload after I accidentally chopped down my trees ahead of schedule.

In fact, I'm really learning to save often - I also nearly missed my chance to stop work on the Barracks in my capital before it was done. If I'm having trouble watching 2 cities, I'm in deep doo-doo once I expand...

Unfortunately, the Civ4lerts mod doesn't seem to be working. In theory, that is supposed to show a blurb whenever your city is about to have an angry or unhealthy person, or the population is about to go up. With the SE strategy, micro-managing is important so it'd be really nice to have...

My research has churned out Bronze Working, Meditation, and Priesthood (just this turn).

Here's my current situation:


Not bad. I'd like to snag those Horses and Copper quickstyle so I can shift to offense.

Next Steps
1. Reseach Pottery, Masonry
2. City 1: queue Oracle after Barracks, whip-finish Barracks to give overflow to Oracle
3. Continue pre-chops and roads
4. Build 1 mine in each city to boost hammers (if time)
5. Chop down forests around Oracle to finish
6. Build Warriors while waiting for research to complete
7. Use Oracle bonus to get Metal Casting

SE1: to 3000BC

The first big event was a goodie hut with a Scout in it. Sweet. Followed by a second goodie hut for 46 gold. Then a 3rd goodie hut which gave him XP, and he's now a speedster woodsman blazing through the northern jungles, trying to find more huts and enemy civs.


Reseach-wise, I think I've made good progress:
3680 BC: Agriculture
3280 BC: Animal Husbandry
3000 BC: The Wheel.
I'm now on track to get Bronze Working shortly.

On the diplomatic front, I found my neighbors. Mehmed II is on my west, while Peter is somewhere to the distant north or northwest. This will be interesting. I think both are moderate but not great trading partners, and both are moderately but not exceedingly belligerent. I will try to court Peter and focus on crushing Mehmed as he's the closer threat.

The geography is pretty good. I appear to be in the lower corner of a continent with a big jungle barrier to the north. I can use that jungle as a buffer as I consolidate the south.

Most importantly, how is progress on the gambit? Well, I'm not totally stoked about options for my 2nd city. If you look below, I'm leaning toward the space directly NE of my capital. It'll overlap on a couple forests, and has some +3 Food squares. I think it'll work. My Fast Worker has finished up making a Farm on the Wheat and a Pasture on the Pigs and is now making roads in the NE Forests.



Next Steps
1. Finish Settler
2. Worker builds roads
3. Settle NE site
4. City #1 begins a Barracks but stops just 1 turn shy of finishing
5. City #2 begins a Worker
6. Workers pre-chop 3 Forests pre city (esp. around City #2) but stop 1 turn shy of finishing
7. Research Meditation
8. Research Priesthood

Specialist Economy attempt 1: Pre-game

And now for something completely different. There's a crazy way to win at Civ4 where your research slider is at 0% for most of the game. All your research comes from scientists. It's called the Specialist Economy and much info on it can be found at the CivFanatics website.

This as totally piqued my interest so I'm going to try it, preferably using the equally intriguing Oracle/Metal Casting/Pyramids Gambit. This gambit aims to get a Forge super-quick, use an Engineer specialist to then get a Great Engineer, use him to build the Pyramids, and then use the Pyramids to get Representation really early. To pull this off requires ridiculous timing and min/maxy sacrifices with Slavery and pre-chopping of forests.

Game settings
Version: Warlords v2.13
Difficulty: Noble
Game Speed: Normal
Map: Continents
Size: Standard
Climate: Temperate
Sealevel: Medium
Era: Ancient
Civ: Indian
Leader: Gandhi

To help with micro-managing, I'm also trying out the Civ4lerts mod.

I choose Gandhi because the starting techs are perfect for these strategies. Philosophical appears necessary for a Specialist Economy (SE). I plan on switching Civics a lot to maximize timing of Theocracy/Vassalage benefits for war.

Here's a screenshot of my starting position:


I think that will work. I have lots of 3 Food tiles, lots of Forests, and it appears that I can position my second city on the opposite side of one of those forest clumps.

Here's my initial plan:
1. Research Agriculture
2. Queue & Build Worker
3. Explore
4. Set Worker to prepare gambit
5. Queue & Build Settler

Next post will cover how it all goes...