Staking Plans and Hockey Betting Bankroll Management

Tracking Drawdowns and Flat Staking Systems

My first NHL season ended with a roughly break-even profit-and-loss line, which I treated at the time as a moral victory. The truth was uglier. I had won on a handful of correctly-read playoff series and lost steadily across hundreds of small regular-season bets, and the only reason I had finished anywhere near even was that one Cup futures ticket had carried the whole campaign. I had no idea, going into the next season, whether I had been profitable on moneylines, profitable on totals, profitable in-play or profitable on futures, because I had not been tracking. I just had a bank balance and a vague feeling.

The spreadsheet I built that summer is the foundation of every betting decision I have made since. This article is what is in it, and how to build your own.

What a bankroll actually is

Your bankroll is the money you have allocated specifically for betting, separate from your day-to-day finances. It is not your savings. It is not money you might need next month. It is money you have set aside knowing it could disappear, and the disappearance would not change the rest of your life materially.

The mistake most bettors make is treating the betting account balance as the bankroll. They are not the same thing. Your bankroll is the total pot dedicated to betting, which can sit across multiple operator accounts plus whatever is in your linked debit card waiting to be deposited. The total is the number that matters for sizing decisions, not the balance at any single book.

The honest test for the size of your bankroll: if it dropped 50 % over the course of one bad month, would your rent get paid on time? Would your partner notice? Would you miss any planned spending? If the answers to all three are “no, no, no”, the bankroll is appropriately sized. If any of them is “yes”, reduce the bankroll until they are all “no”. Bankroll management starts with bankroll sizing.

The unit system

The bedrock of bankroll management is the unit system. A “unit” is a percentage of the bankroll — typically one per cent — that represents the standard stake size for a single bet. A bettor with a £2,000 bankroll using a 1 % unit stakes £20 on a standard bet. A bettor with a £10,000 bankroll using the same 1 % unit stakes £100.

The unit system does two things. First, it scales the stake automatically with the bankroll. If you win, the unit grows. If you lose, the unit shrinks. The growth or shrinkage is gradual and never threatens the survival of the bankroll, because the percentage stays constant while the absolute number adjusts. Second, the unit removes the emotional weight from any single bet. A £20 stake feels like a £20 stake regardless of whether it represents 1 % of a stable bankroll or 1 % of a bankroll that has just doubled. The mental anchor is the unit, not the absolute pounds.

Professional bettors typically nail down 2 to 5 % edge per bet using fractional Kelly and flat percentages of bankroll. The “flat percentage” is the unit system in practice. The Kelly adjustment scales the unit up or down based on the strength of the edge, but the base unit is the anchor.

Sizing units by confidence

Not every bet deserves a 1-unit stake. The simple version of confidence-based sizing uses three tiers: 0.5 units for marginal edges where you have only modest conviction, 1 unit for standard bets where the model and the intuition align, and 2 units for high-conviction bets where the price is well off your model and the matchup is unusually well-understood.

The hard discipline is keeping the high-conviction tier rare. Most bettors find themselves rating roughly 30 % of bets as high-conviction, which is far too many. The honest number is closer to 5 to 10 %. If more than one bet in ten is in the 2-unit tier, the system is broken and the bettor is rationalising stakes upward to chase the action.

My own discipline: I do not place more than two 2-unit bets in any single calendar week, and I keep a running log of how those high-conviction bets perform versus my standard 1-unit bets. If the win rate on 2-unit bets is not at least three percentage points higher than on 1-unit bets across a season, the tiering is not adding value and the system needs recalibration.

What the spreadsheet actually tracks

The columns in my spreadsheet, in order: date; sport (NHL by default); market type (moneyline, puck line, total, prop, future, in-play); selection; price taken (decimal); stake (in units and in pounds); closing line at game start; result; profit/loss in pounds; profit/loss in units; closing-line value calculated automatically; cumulative bankroll; notes.

The most underrated of those columns is closing-line value. CLV is the gap between the price you took and the closing price at game start, expressed as a percentage. A positive CLV across many bets is a more reliable signal of long-term profitability than a positive P&L over the same sample. Bettors with positive P&L and negative CLV are running good. Bettors with positive CLV and negative P&L are running bad. Over enough bets, CLV catches up to P&L. Run-good and run-bad average out.

The notes column captures the reasoning — why the bet was placed, what the edge was, what could go wrong. The point is not to write a novel; it is to leave enough of a trail that a year later, reviewing the season, you can see which categories of bet actually paid off and which were rationalised but unprofitable.

Drawdown tolerance and the survival question

Every bettor will experience a drawdown. The mathematics is essentially unavoidable: even a long-term profitable strategy can produce stretches of 15-30 losing bets in a row, and the bankroll will swing accordingly. The question is whether the drawdown is survivable.

With a 1 % flat unit, a 30 % bankroll drawdown is psychologically painful but mathematically modest. With a 5 % unit, the same drawdown sequence can wipe out the bankroll entirely. The unit size dictates the survival probability of the strategy, not the win rate.

UK Gambling Commission data underscores the broader point — the UK gambling industry remains one of the largest regulated betting markets in Europe, and the volume of customer activity across UK-licensed operators means that drawdowns at the individual customer level are statistically certain even for skilled bettors. The licensed regulator has been the sole authority for British gambling licensing since 2007 and continues to enforce affordability and harm-prevention standards across the industry. The system the regulator polices does not protect bettors from drawdowns — it protects them from the worst behavioural consequences of drawdowns, by requiring deposit limits and self-exclusion tools across every licensed operator.

The bettor’s job is to set the unit small enough that no realistic drawdown threatens the bankroll. My personal rule of thumb: a 50 % drawdown should be possible without forcing the bankroll below the minimum stake of any market I want to bet. That implies, for typical UK NHL market minimums of £0.20 to £1, a starting bankroll of at least £100 per unit — i.e., at least £10,000 bankroll for a 1 % unit of £100. Most punters bet smaller than this. Most punters use unit sizes that look fine in good months and become catastrophic in bad ones.

The withdrawal habit

The single best behavioural discipline I have learned is the periodic withdrawal. Every time the bankroll grows by a defined percentage — say 25 % — I withdraw half the gain to my bank account. The other half stays in the bankroll to compound. The withdrawn money is no longer part of the betting system. It cannot be re-bet, cannot be chased, cannot be lost.

The reason this matters psychologically is that betting bankrolls have a tendency to feel weightless. The money in the betting account does not feel like the money in the current account. The withdrawal habit forces a regular conversion of betting wins into real-money outcomes, which anchors the betting activity to the actual finances rather than to an abstract running total.

The reason it matters mathematically is that every withdrawal is a guaranteed lock-in of profit. The bankroll that survives the season is the one that has been pruned periodically rather than the one that has grown unchecked. The bettor who never withdraws is the bettor who is one bad month from giving everything back.

What the system buys you

The bankroll system above — sized appropriately, unit-based, confidence-tiered, tracked obsessively, withdrawn periodically — does not make you a winning bettor. Winning bets do that. The system protects the bankroll from the inevitable losing bets, smooths the volatility into something psychologically survivable, and produces the data you need to actually identify which of your strategies are working.

I have run roughly the same system for years now, with minor refinements. The bankroll has grown. The drawdowns have been recoverable. The spreadsheet tells me which markets have paid and which have not, which lets the next season’s strategy adjust accordingly. None of this is glamorous. All of it is the foundation on which the more interesting strategy questions — closing line value, Kelly sizing, market selection — can actually be answered.

For the natural next step on turning the bankroll into a strategy with measurable edge, the expected value in ice hockey betting piece is the companion.

Written by the editors at Ice Hockey Betting Tips.