Skip to content
Television Radar

Updated 6 Nov, 2026

How we calculate the score

The spec sheet becomes a score from 0 to 10, with public weights.

What counts toward the score

The score is the average of 6 components, each from 0 to 10. If the spec sheet doesn’t list one of them, it drops out of the math and the weights of the others are recalculated; with fewer than four components, or without the picture component, the TV gets no score. Size is left out because bigger is not better: the right size depends on how far the couch is. The weights are image 40%, gaming 20%, system 15%, connections 10%, audio 10% e recency 5%.

  • Image40%

    Panel type (OLED, Mini LED, QLED, standard LED), resolution, HDR and, when the spec sheet lists it, brightness

  • Gaming20%

    Native refresh rate, HDMI 2.1 and VRR; without the native rate on the spec sheet, the TV gets no score here

  • System15%

    Operating system (Google TV, webOS, Tizen, Roku...) and confirmed Wi-Fi and Bluetooth

  • Connections10%

    Number of HDMI ports, eARC and USB ports

  • Audio10%

    Stated power, number of speakers and Dolby Atmos

  • Recency5%

    Model year: Televisions get a new model every year, so older ones lose points

What the score doesn’t consider

  • Measured contrast and peak brightnessThe spec sheet lists the panel type, not the black level the TV delivers or its real brightness. Peak brightness is rarely on the spec sheet.
  • Calibration, uniformity and viewing angleTwo Televisions of the same model leave the factory with different colour and light settings, and the viewing angle is not on the spec sheet.
  • Room, light and glareA bright room calls for brightness and a screen with little reflection. The spec sheet doesn’t know what your room is like.
  • Input lag and motion picture qualityNative refresh rate counts; input lag, upscaling and colour accuracy are not on the spec sheet.
  • Durability and software after two yearsHow long the TV lasts and how long the system gets updates is not on the spec sheet.
  • Size, brand and power useSize is not part of the score because bigger is not better: the right size depends on how far the couch is. Brand and energy use are left out too.

Two Televisions with similar spec sheets can perform differently. That is why the score is for comparing Televisions of the same size and at the same price level and finding where paying more stops paying off, not for choosing on its own.

Next step

See the score of each TV next to the price

The chart plots all the new TVs by price and score.

See the TVs on the chartCompare two TVs

How to read the score

It measures what the spec sheet promises. A 4K TV with an OLED or Mini LED panel, native 120 Hz, HDMI 2.1 and a recent release lands among the highest scores. An entry-level Full HD TV with a regular LED panel and 60 Hz lands among the lowest. The same math applies to every brand.

Two details that surprise people: 8K barely gains points over 4K, because there is little 8K content to watch; and a TV that doesn’t list its native refresh rate gets no gaming score, without losing points on the other components.

Score from 0 to 10 calculated by the Radar from the TV’s spec sheet (image, gaming, system, connections, audio e recency). It does not measure real contrast, peak brightness, calibration, input lag, durability or what buyers think. Size is not part of the score.

What does each axis of the chart show?

The chart always puts price on the horizontal axis. On the vertical axis you pick the question:

  • Spec sheet: the 0 to 10 score on this page.
  • Imagem: only the picture component (panel, resolution, HDR and brightness), from 0 to 10. It is a spec sheet score, not a lab measurement.
  • Games: only the gaming component (native refresh rate, HDMI 2.1 and VRR). A TV that doesn’t list the native rate is left out.
  • Cheap for the specs: how far the price is below the median of Televisions with a similar score and size (up to 0.5 points of score apart, same inch size or the neighboring one in the size list, at least 5 Televisions). Only trusted sellers are included, with no price jump of 20% or more between updates, and only cases where a price curve by score and size also points to a below-expected price. The score comes from the spec sheet, so brand and software are not part of the math.
  • Today’s finds: how far today’s price is below the median of recent days from the same seller. It only appears when there are at least 7 data points in the history, and a one-day drop only counts with a trusted seller or when confirmed in the next collection.

The line on the chart connects the Televisions that no other TV beats on both price and the chosen axis. On the price difference axes there is no line: the biggest differences get labeled.

Where does the data come from?

The attributes come from the spec sheet on the Amazon product page and, when the model is the same, from a second catalog; each data point keeps the source it came from. The price is the lowest among trusted sellers (the store itself or a seller rated 4.5 with 100 or more sales) and carries the update date. The score’s source appears on screen next to each TV. If we start using another source, this page and the screen say which one and since when.

How to read these numbers

The score goes from 0 to 10 and comes from the TV’s spec sheet: image, gaming, system, connections, audio e recency. The math is the same for every model and is on this page. It says what the TV delivers on paper, not how it performs in your living room. Size is not part of the score.

Still have a question?

Write to contact@radarofproducts.com.