Players who can see a sort-direction arrow on a lobby filter cut their time-to-result by roughly 4x compared to a list with no visible indicator. That's the headline finding from a set of usability sessions run across three markets in late 2024, and it holds up whether the player is scanning 40 slots or 4,000 sports markets. The arrow itself does almost nothing. What it changes is whether the player believes the list has already been sorted.

The arrow is a promise, not a control

In the sessions, testers were given two versions of the same lobby. Version A had a clickable column header with no icon. Version B had the same header plus a small chevron pointing down. Same data, same sort logic, same latency.

Median time to find a target game dropped from 11.4 seconds to 2.9 seconds on Version B. That's the 4x. The interesting part is why: almost nobody clicked the chevron. They just read the list as already ordered and started scanning from the top.

When there's no indicator, players assume the list is in some arbitrary or promotional order. So they don't scan — they read, top to bottom, checking each entry. That's a fundamentally slower behaviour, and it's the default state on a surprising number of operator lobbies.

What the eye actually does

Heatmaps from the same sessions showed Version A users making 6–8 fixation points per screen. Version B users made 2–3. The arrow doesn't add information about the games. It removes the question of whether to trust the order.

Where operators get this wrong

Three patterns showed up repeatedly across the lobbies we looked at:

• Sort arrow appears only on hover. On desktop this is survivable. On touch it's invisible until after the tap, which means the player has already committed to a scroll.
• Arrow shows direction but not the active field. A down-chevron on "Popular" and a down-chevron on "New" look identical. Players can't tell which sort is live.
• Sort resets on filter change. Player sorts by RTP, applies a provider filter, and the list silently reverts to default order. The arrow still points down. Now it's lying.

That last one is the expensive mistake. Once a player catches the interface contradicting itself, they stop trusting the order entirely and go back to linear scanning — the exact behaviour the arrow was meant to prevent.

The mobile problem is worse

On mobile, the gap between sorted and unsorted lobbies widened to about 5x in the same tests. Smaller viewports mean fewer items visible per screen, so a linear scan costs more scrolls. Anything that lets a player skip the scan pays back faster.

The practical fix is unglamorous: a persistent sort chip above the grid showing the active field and direction ("RTP ↓"), tappable to change. No hover states, no hidden menus. It costs maybe 40 pixels of vertical space.

Does the 4x survive real money?

Here's the open question. These were lab sessions with no stakes. A player browsing for fun has different patience than someone with a live accumulator and 90 seconds until kickoff. It's plausible the arrow matters more under time pressure — but it's equally plausible that experienced players have already built mental shortcuts and the indicator adds little.

Nobody has published retention or session-length data tied to sort-indicator changes, which is the number that would actually settle it. Until someone does, treat the 4x as a strong signal about interface trust, not a guaranteed conversion lift. And if you're running these tests yourself, watch the players who never touch the filter at all — they're the ones the arrow is really for.