Steam sale strategy, not just discount data

Steam deals are easy to find.
Good buys are harder.

ATLGames tracks price history and all-time lows, but the point of the site is not to dump a wall of discounts on you. It is to help you decide whether a sale is actually good, whether you should wait for a better window, and when a "great deal" is really just backlog bait.

Live data

Separate deals browser

The live discount feed lives on its own page so you can filter by price, genre, release status, and all-time-low flags without turning the homepage into a giant list page.

Original writing

Editorial buying strategy

The guide and blog focus on timing, price history, bundles, and backlog discipline so the numbers are backed by actual decision-making context.

Practical rules

Use the floor, not the sticker

A 75% off tag says less than the gap between today's price and the historical low. That is the lens the rest of the site is built around.

What makes a sale actually worth buying?

Start with the historical floor. If a game is still well above its best recorded price, the discount percentage is mostly decoration. After that, look at frequency. A game that hits the same low every few months gives you no reason to rush.

Then be honest about your own timing. A technically great price is still a weak purchase if you are buying for some imaginary future version of yourself who will magically clear their backlog. That is why the site now splits the surfaces: the homepage is for context and methodology, and the deals browser is for the raw listings.

How we identify a real deal

Our four-step methodology

Every deal listed on ATLGames.io is evaluated against a deterministic price-history model before it shows up in the feed. This is the same logic that powers the per-game analysis on every game detail page. We publish it openly so you can decide whether our recommendations match the way you actually buy games.

  1. 01

    Compare against the all-time low (ATL), not the MSRP

    Discount percentages are calculated off of an inflated full price, so a “75% off” badge on a game that frequently goes 80% off is a worse deal than a “40% off” sticker on a game that almost never discounts. We compare today’s price to the historical floor across every store we track and tag deals that are at, tied with, or within 10% of that floor. Anything else is just "a sale".

  2. 02

    Measure how often the game reaches that floor

    Some titles hit their lowest price every seasonal sale; others reach it once and never again. We count distinct sale events and the average gap between them. A game whose floor recurs every three months is a low-urgency buy. A game that only hit its floor twice in two years is a much stronger “buy now” signal at the same headline discount.

  3. 03

    Account for release age and DLC dilution

    A 30% discount on a six-week-old AAA release is structurally different from a 30% discount on a four-year-old game whose complete edition exists. We surface release date alongside the deal and call out when a base-game sale is likely to be superseded by an upcoming bundle. Our DLC math guidewalks through the actual decision tree.

  4. 04

    Filter out deals that aren’t actually deals

    The optional “hide bad deals” filter removes anything currently priced more than 50% above its historical low, regardless of the headline percentage. We also exclude unreleased games and DLC by default so the list reflects the kind of purchases people actually make on impulse. You can re-enable both filters on the deals browser.

We are deliberately conservative about flagging things as “buy now”. If a deal isn’t at the floor or within reach of it, the site is more likely to tell you to wait. That bias is intentional — most Steam buyers already own more games than they’ll ever play, and another mid-tier discount doesn’t change that math.