In team Blood Bowl, playing optimally for yourself rather than the team can be the fastest way to lose the round.
TLDR
- Round Over Board: Optimise for the team’s total points rather than individual win rates; draws are often highly valuable.
- Variance Roles: Manage risk using Anchors (stability), Aggro (intentional volatility), and Pivots (tactical flexibility).
- Threshold Discipline: Once a round is secured, suppress further risk as additional wins provide diminishing utility.
- Meta-Agent Coordination: Captains synchronise board strategies to align individual play with the team’s collective success.
From Zero‑Sum to Non‑Zero‑Sum Thinking
In a standard singles match, the structure is largely zero‑sum. One player’s success directly corresponds to the other’s failure. Team tournaments fundamentally alter this landscape. Individual games are no longer the final object of optimisation. Instead, they are components of a larger system whose objective is winning rounds. This introduces non‑zero‑sum dynamics, where individual draws can contribute positively to the collective outcome depending on context.
The most common failure when players transition into team formats is continuing to evaluate decisions through an individual lens. Game‑theory clarity comes from recognising that the utility of an outcome is defined at the team level, not the board level. Once this shift is internalised, decisions that would feel overly cautious in singles play become rational or optimal within the team framework.
Portfolio Thinking Across Boards
A team round can be understood as a portfolio of independent but partially correlated outcome distributions. Each board contributes a distribution of possible results, and the aggregation of these distributions determines the likelihood of winning the round.
Round success does not require maximising every board’s expected value. We don’t all need to win. Winning the rounds (consistently) requires shaping the combined distribution so that enough probability mass lies on winning round outcomes.
This perspective reframes familiar behaviours. Multiple players independently injecting variance (trying to win from sub-optimal positions) may increase the chance of a round win, but also greatly increase the chance of a loss. Conversely, deliberately suppressing variance on advantaged boards while concentrating risk in disadvantaged matchups produces tighter, more reliable round‑level outcomes. At least, that’s the plan.
Threshold Games and Diminishing Returns
Team tournaments are usually threshold games: it’s only after you’ve won all your rounds that tiebreakers need to be considered. Therefore, once a team has achieved enough points to secure the round, further upside has diminishing utility while additional risk introduces new loss modes. This fundamentally changes optimal behaviour late in rounds.
To see this in action you only need to look at the Tilean Cup. The winning team had fewer wins and more losses than the teams in 2nd and 3rd, and even a slightly worse overall record than 4th place. They won the games that mattered. We’ll come back to the data from this tournament shortly.
Failure to recognise thresholds produces a potentially costly error: continuing to push for more wins (and increasing the risk of losses) after the round is effectively secured. Aggressive plays, or attempts to convert draws into wins may feel productive but simply adds variance that only benefits the opposing team. Conversely, recognising that the threshold has been crossed allows teams to intentionally shut games down, protecting the collective result even when individual positions could be pressed further.
Coordination, Communication, and Trust
Multi‑agent games introduce coordination problems that do not exist in singles play. The rational strategy for each individual depends on credible expectations about what teammates will do. Without communication, players default to personal optimisation, even when it harms the team. Effective teams reduce this problem through clear role assignment, shared language around risk, and mutual confidence that restraint will not be punished socially.
This coordination requirement also explains why team Blood Bowl can feel qualitatively harder than singles, while simultaneously is more rewarding.
Decisions can no longer be made in isolation. They must account for partial information about other boards and evolving round dynamics. When coordination works well, it reduces cognitive load rather than increasing it, because players can focus on executing their assigned role rather than re‑evaluating global objectives every turn.
Indeed, in a Game Theory thinking you could treat the captain as a meta‑agent managing the team’s aggregate expected value; a strategic layer above individual boards. For this reason, many of the top International teams now include a non-playing team member to provide that coordination and communication.
Team play also introduces new psychological pressure. Certainly, when I play in a team I feel social pressure bias, I don’t want to be the one who “cost the round.”
Caring about the results is fine in principle, but can be a problem if this distorts risk tolerance. If your role in the team (more on this below) is to be one of the points where variance is supposed to be injected, then with that role comes increased risk of losing. Strong captains and teammates will mitigate social pressure bias by reframing responsibility as collective rather than individual.
We win and lose together. Or, as Scotland rugby say #AsOne.
Anchor, Pivot and Aggro Boards
Once we accept that not everyone needs to always play for the win, we can appreciate that within the squad not every board needs to serve the same strategic function. Each matchup contributes differently to the round‑level expected value, and understanding those roles is essential to building a coherent squad.
We can think about roles as three archetypes: anchor boards, pivot boards, and aggro boards. Together they form the team’s variance management system; a way of distributing risk and reliability across 6/8 (at World Cup/Eurobowl respectively) games so that the collective outcome becomes predictable.
Anchor boards are played with low‑variance races and generally conservative playstyles, designed to avoid losses as the primary objective. Their purpose is to secure the team’s floor: to ensure that even if other boards collapse, the round remains recoverable. In a four-player tournament, you might forgo anchor boards with greater emphasis on pivot boards.
Pivot boards sit in the middle of the distribution. These races and the coaches playing them should have flexibility; capable of switching to push for wins if the round demands it, but also low variance enough to lock down draws when others in the squad are ahead.
Aggro boards are the pre-planned variance insertion points. These are higher‑variance races deliberately positioned to create opportunities and push for wins. An aggro board should bring controlled volatility. Against teams with higher average coach skill levels, these boards will look to steal the round while the rest of the squad aim to secure draws.
Assigning coaches to their core roles involves aligning player skill, playstyle and racial preference with variance exposure needs. Not easy. It also involves self awareness… So, let’s get introspective.
Reliable, methodical players playing steady races are best suited on anchor boards, where patience and precision prevent collapse. Players who hate losing more than love winning. Perhaps the people who are known for playing tight consistent BB, or perhaps are more toward the conservative side on their personal risk profile. Adaptive thinkers playing races that can shift between modes belong on pivot boards, where flexibility and communication are paramount. The most creative and resilient players, those most comfortable with risk, should pilot the aggro boards with races able to inject variance.
Most races straddle more than one role. This means Captains can flex to adjust emphasis dynamically in response to racial and player matchups.
These assignments should be explicit. Captains communicate not only which race each player will run but why that race fits their role within the team’s variance structure; generally and for that specific round. Clarity here transforms eight independent games into a coordinated system. Clarity also serves to reduce cognitive load. Everything should have a clear view on what they are doing and so can focus on the game.
Building a Squad – Team Scotland Example
Let’s shift from the abstract and use the specific example of Team Scotland. This is not (just) because I’m a member of the team but because we are a mid-tier nation (currently) with aspirations to improve our ranking. We finished 14th at Eurobowl 2025, good enough to secure hosting rights for EB2028… come see us in Glasgow! For a nation like Scotland aiming for a Top‑10 (of ~30) finish is the next step on our upward journey.
Increasing our ranking means that our strategy should balance being able to play low variance blood bowl vs mid and low tier nations where we should expect to win, while retaining enough opportunity to flex and increase variance where needed to steal surprise wins against stronger opponents. We have a finite pool of coaches available but there is competition for places. Also, unlike some of the bigger player-base nations, we don’t necessarily have complete coverage of race-specific specialists, so we need options. What does that mean for race choices?
From a strategic perspective we want to increase variance against top tier nations, suppress variance in rounds we should expect to win and have a balanced approach against teams at a similar level to us. We need a racial composition that allows us to switch between these modes.
This is where population‑level data becomes useful. I’m going to use two sources here. FUMMBL games played in the last 30 days using EB2026 rules, and the Tilean Team Cup. Although the FUMMBL dataset mixes elite coaches with beginners, the shape of the bars still reveals something fundamental: the intrinsic stability of each race. Races with low loss rates across the entire population are inherently forgiving; races with wide spreads are inherently volatile. For the Tilean Cup we’re going to look at the squads toward the top of the table.
The comments below are to illustrate the thought process rather than necessarily the answer. Moreover, the meta might shake out further in the next few months.
When you look at the FUMBBL data, a clear pattern emerges. The races with the lowest loss rates are Shambling Undead, High Elves, Humans, Dark Elves, Necromantic, and Lizardmen. This forms a natural “core.” The Undead, Necro, Lizards and Humans could adopt the anchor or pivot roles in any particular matchup, whereas High and Dark Elves can naturally sit in the aggro or pivot positions. Khorne are also doing well, but with a low number of matched played.
What about the final boards? The final 2-3 should be selected not just by raw win rate but by function and meta. The FUMBBL data provides a baseline for racial stability, but the top tournament meta determines contextual value. A race with moderate variance can become disproportionately powerful if it counters what’s popular.
So let’s look at what the top teams are playing? The Tilean Team cup is a 4 player tournament rather than 8, but those data give us an idea of what the perceived “must haves” are, and a measure of how successful those teams can be. I looked at the composition of the top 15 (of 72) squads. In this list are stalwarts of Team England, Italy, Spain, France, Germany, Sweden (and BB_jock representing Scotland).
High Elves (11) and Dark Elves (9) dominated. Necro and Nurgle (4 each), Human, OWA and Elven Union (3 each) were next with WE, Lizards, TKs, Nobs and Zons (2) and a smattering of other teams selected once.
For the final boards, Scotland could select a high‑variance high win-rate race (Wood Elves, EU) in the hands of a top coach. This would give the squad an additional fully-committed aggro board, capable of converting difficult pairings into unexpected wins… but it might also inject too much variance? Conversely, adding an additional anchor (Nurgle, Orcs) could provide more stability across the 8. This would put more emphasis on the other aggro/pivot boards to grab wins. Alternatively, a meta‑counter race could fill a pivot position like Amazons or OWA* this might give the team more round-specific flex. The final decision isn’t easy.
These final choices are about shaping the team’s variance curve, not chasing individual EV.
What do you think?
There isn’t an objectively right option. The blend of stability with aggro specialists is like loads of other aspects of BB, a compromise where shifting in one direction makes the team more likely to win vs top 10 teams but more at risk of a surprise loss vs bottom 10, whereas shifting the other way would increase the win rate in the games we would expect to win but perhaps lower the win-rate vs to the elite.
Overall the goal is to find the optimal balance that produces a round‑level distribution that is narrow enough to avoid collapse yet flexible enough to generate upside when needed. Tricky. And then the players need to execute.
Using Game Theory to Improve your Team’s Results.
Train yourself to judge decisions based on their contribution to the round outcome rather than the match result alone. After team games, review moments where personal optimisation conflicted with team interests and evaluate whether restraint would have increased round win probability. Also, learn to love a draw!
Practise identifying your role within the team portfolio each round, whether that role is to stabilise, to absorb risk, or to remain flexible. In team discussions, focus less on who “should win” and more on where variance should live.
Work on communicating your game state clearly and concisely to teammates, especially around stability and risk. Practise trusting agreed team strategy even when it conflicts with your instincts. Improvement shows when team discussions are proactive rather than reactive, and when individual players feel comfortable accepting results that serve the collective goal.
Post‑Round Review: Encourage teams to analyse not just board outcomes but risk allocation efficiency: did variance concentrate where it should? Did any player over‑optimise for personal win rate?
Leave a Reply