Format First, Numbers Later: The Basic Rule of Cricket Analysis Everyone Forgets

Cricket is among the most heavily measured sports in the world. Every ball, every shot and every over generates data, and analysis starts circulating on social media before the match is even over. Yet in that flood of numbers, one basic step is routinely skipped: before deciding what any statistic means, you must decide which format you are talking about. Test, ODI and T20 are three different games under one name. Their tactical logic differs, the purpose of pitch preparation differs, and the interpretation of every performance indicator differs. A number that proves excellence in one format can prove below-average output in another. In Test cricket the scarcest resource is time. Five days, four innings, and a game in which a draw is a legitimate result. Speed is not the primary virtue here; patience, wearing the ball down, and not losing wickets in a session are the real issues. A batter averaging 45 at a strike rate of 50 in Tests is invaluable to his team, but place the same numbers on a T20 scale and that same batter becomes a liability. Conversely, a finisher with a strike rate of 150 can walk back to the pavilion in three balls in a Test. So whenever a number appears, the first question must be: which format? Second: what is the sample size? Third: is this a home-ground number or an away number? Phases within a match matter just as much. In limited-overs cricket the first six overs, the middle overs and the last ten are three different games. In the powerplay both wickets and runs carry a high price; in the middle overs spinners take charge and the run rate often stalls; and in the death overs a 20-run over or a two-wicket over can flip the entire result. A bowler conceding at 8 an over in the death phase cannot be read the same way as an economy of 8 in the powerplay, because the pressure is different. In Tests the same idea surfaces as sessions. The morning session, the afternoon session and the evening session behave differently as the pitch changes, so a generic figure like '400 in the first innings' says nothing on its own. At the root of every analysis are small units of fact — a scoreline, a bowler's figures, a quotation, a contract amount. These are information points. The rule is simple: every conclusion must trace back to at least one verifiable information point. If a claim has no solid fact behind it, it is not analysis but conjecture. Presenting conjecture as analysis is journalism's biggest weakness, because readers take it as fact. Venue and environment are the next filter. The same team's home and away records are often two different stories. A score made on a subcontinental pitch built for spin and a score made on a bouncy Australian or English surface cannot be weighed on the same scale. A target changed by dew, a target revised under DLS after rain, wind direction and cloud cover affect results as much as any shot. This does not mean using them as excuses; it means isolating these factors before calling a performance good or bad. Toss luck and DLS arithmetic are elements that should be stripped out explicitly, otherwise we end up calling skill by the name of fortune. Player evaluation is the most delicate part. The first task is to fix the role — opener, anchor, finisher, fast bowler, spinner, all-rounder, wicketkeeper-batter. To take examples, Jasprit Bumrah operates as a fast bowler, Ravichandran Ashwin as an off-spinner, Joe Root and Steve Smith as Test batters, Babar Azam as a top-order batter, Rashid Khan as a T20 specialist leg-spinner, Shai Hope as a wicketkeeper-batter, and Hardik Pandya as a pace-bowling all-rounder. Only after the role is fixed can you judge whether a number is low or high for that role. A finisher's strike rate and an anchor's strike rate cannot be judged on the same standard, because the team expects different things from each. The second task is comparing career averages with the last twelve months. If a player has averaged 50 over his career but 38 in the past year, that decline is a signal — but it is not yet proof that it is permanent. He may have come off a tough tour, or been run out two or three times. This is where sample size matters. Judging a career on ten innings is like forecasting the weather from one rainy day. The third task is the age curve: fast bowlers peak in pace and endurance around thirty and decline gradually after that, while batters mature later. The fourth factor is injury history — invisible in the data, yet the biggest hidden variable in performance. In assessing teams, the home-away differential is the single largest driver. Hardly any other variable in international cricket is as large. Next comes squad structure — batting depth, bowling combination, bench strength and age profile. A side with six bowling options plays under different pressure from one with only four. Bench depth also matters beyond injuries: it manages selection politics. Teams with options can rest players; teams without them grind the same eleven through a season. ICC rankings offer a reference, but they vary by format and cannot be read directly as proof of 'best team'. Leagues and commercial ecosystems follow a different logic. The essential separation here is that a player's market price and his on-field utility are not the same thing. There are four common reasons for a high auction bid: demand for a local young star, a shortage of all-rounders, scarcity of a specific role, and finally panic bidding. The Right to Match provision lets a former franchise match the highest bid, which changes the entire auction arithmetic. Broadcast-rights value, franchise valuation and player salaries are all indicators of a league's financial health, not of its sporting quality. Conflating the two is the most common analytical error. Rules and governance are often ignored, yet the biggest decisions come from there. An NOC permitting a player to appear in an overseas league, or the terms of a central contract, directly determine player availability. Anti-corruption monitoring rests with the ICC's Anti-Corruption Unit, and when a case surfaces the questions fall not just on the player but on the whole system. The World Test Championship gave Test cricket a points table, changing the weight of every series — no Test is 'just a Test' anymore. Assessing a team's strategy without understanding these institutional shifts is incomplete. And finally, narrative versus data. Cricket builds stories fast — two brilliant innings make a young player 'the next superstar', three failures declare a veteran's 'end'. Neither conclusion survives the sample-size test. Form built in a short league burst and multi-year sustained output cannot sit on the same scale. The gap between expectation and reality — commonly called hype — becomes visible only when we ask how much information sits behind the claim. And that question should be asked of every claim, about a player or a team. The risk side deserves a look too. In a crowded calendar the biggest risk is workload and the injuries that follow, especially for fast bowlers who must switch formats at short notice. The second is commercial — the broadcast-rights price cycle and the financial fragility of smaller boards. The third is credibility and governance risk, where match-fixing or suspension affects the credibility of the entire sport. The fourth is climate and geopolitics — extreme heat, rain and political tension can question the staging of matches themselves. None of these risks is isolated; an injury lowers a player's auction price, which in turn changes team strategy. It is also worth remembering that any risk assessment only makes sense when it has a subject — a team, a player, a series. Saying 'risk is low' without a subject is the biggest mistake of all, because unknown risk and low risk are not the same thing. The sum of this argument is that cricket analysis is not the art of forming opinions but the discipline of asking questions. The first question is about format, the second about sample size, the third about context and the fourth about the source of the claim. Until all four are answered, treat no conclusion as final. Numbers never lie, but to make them speak we must first decide which game they are talking about.