Updated September 2026  ·  By Michael Rainwater  •  9 min read

The UGA you got into is not the UGA your child is applying to.

Sixty-six years of UGA’s own admissions numbers — including the year it went from below the national average to 280 points above it — and a tool that puts your year against your student’s.

1960  5 in 6 1 in 3 of the people who apply get in
1960  401 31,207 turned away in a single year — 78 times as many
1961  −79 +280 points against the national average SAT. UGA used to be below it.

Fall 1960 against Fall 2025, the most recent class UGA has filed. The class of 2030, announced this spring, runs tighter still: roughly 3 in 10 of the people who applied were admitted. The SAT figure is UGA’s mean against the national mean that year — 53 years of it, 1961 to 2024, with several years UGA published no mean at all. Raw scores cannot be compared across that span, because the test was recentred in 1995 and rebuilt in 2016, but UGA and the national mean were rescaled together, so the gap between them survives both breaks.

010k20k30k40k50k60kDesegregated1961fewer than 4 in 5 got in615 turned awayHOPE begins19937 in 10 got in3,349 turned awayRace removed2001fewer than 3 in 4 got in3,008 turned awayCommon App20212 in 5 got in23,401 turned awayClass of 2030 · announced20263 in 10 got in36,200 turned awayappliedadmitted1960197019801990200020102020

Every year UGA has published, Fall 1960 to Fall 2025, with no gaps. Each Fact Book edition carries a rolling ten-year table, so editions a decade apart overlap and chain into one continuous series; where two editions disagree, the later filing wins. Fall 2005 and Fall 2025 have no Fact Book table and come from UGA’s Common Data Set for those years instead, checked against the men’s and women’s rows that sum to them. Both sources are published by UGA’s Office of Institutional Research. Nothing is interpolated. The red line is applications, the black line is offers of admission, and the shaded band between them is everyone turned away — offers did rise over these sixty-six years, but never close to as fast. The dashed reach at the right is the class of 2030, which UGA has announced but not yet filed; it is drawn because it is the class families are asking about, and excluded from every calculation on this page. The four marks: UGA admitted its first Black students, Hamilton Holmes and Charlayne Hunter, by court order in January 1961; the HOPE Scholarship began in 1993; the Eleventh Circuit struck down UGA’s race-conscious admissions in Johnson v. Board of Regents in 2001; and for the Fall 2021 cycle UGA accepted the Common Application for the first time and made tests optional for that one year. Scores were required again from Fall 2022 and applications stayed high — which suggests the Common App, not test-optional, is what moved the line.

What it takes now.

The middle half of the class, and the median, on each measure.

SAT, enrolled first-yearsFall 2025100012001400160012501420median1330Half the class sits inside the shaded band.
GPA, admitted first-yearsClass of 20303.03.54.04.54.134.4median4.27UGA’s own recalculation, core academic courses only.

Two different bases, on purpose: the SAT figures describe the students who enrolled (UGA’s Common Data Set, Fall 2025); the GPA figures describe the students UGA admitted (Class of 2030), on UGA’s own recalculated scale, which counts core academic courses only and can exceed 4.0. A quarter of the class sits below each shaded band and a quarter above it. These are not cut-offs.

The year you applied.

Or your graduation year, minus four. Nothing is sent anywhere.

Where the admissions rate is heading.

The last ten years and the next ten. Fitted on every year since the series changed regime, then shown from 2014 so the part you are living in is legible.

10%20%30%40%50%60%20162020202420282032PUBLISHEDPROJECTEDannounced21%ADMISSIONS RATEfitted on 25 years, 2001–2025

The black line is what UGA has published; the red is the projection, carried forward from the last published year at the fitted rate, with a 90% prediction interval behind it. The fit uses all 25 years since the 2001 break — the year the series changed regime, which a Chow test puts beyond doubt at F = 30.0 against critical values of 3.15 and 4.98. Across those years the rate has fallen 1.30 points a year (R² 0.850). The filled dot is UGA’s announced class of 2030 at 29.8%, which is not in the fit — it is there as a check, and it already sits below the fitted line. Ten years is a long way to carry any trend, and the width of the band is the honest statement of that: by 2035 it spans most of the plausible range. Read the direction, not the decimal.

One thing none of this can tell you.

What score you needed. The SAT was rescaled in 1995 and rebuilt in 2016, so a 1200 in 1985, in 2005 and today are three different achievements, and no honest line runs across those breaks. Current score targets are on the scores guide.

The working.

How the projection was built, and whether it holds up against a year it was never shown.

How the break was found. Four candidates were fitted to all sixty-six years of applications and compared on information criteria rather than on R², since adding parameters always raises R². Two things are easy to get wrong here and are done properly: a model fitted to the logarithm of applicants is a different object from one fitted to applicants, so the log models carry the Jacobian correction that puts all four on one scale and makes the numbers comparable at all; and the parameter count charges for everything estimated, including the error variance and the break years themselves, which were found by searching every possible split rather than assumed. Without that charge a break always looks free. Lower AIC and BIC are better; k is that parameter count. DW is the Durbin–Watson statistic: 2 means independent residuals, and lower means the errors are correlated year to year. R² is reported within each model’s own scale and is not what chose the winner.

ModelkAICBICDW
Linear, level30.7641318.71325.30.12
Log-linear30.9571162.71169.30.33
Log-linear, break 200160.9781124.11137.20.61
Log-linear, breaks 1969 + 200190.9871096.71116.41.00

The two-break model wins on both criteria even after paying for its extra parameters and its searched break years, and the 2001 break it finds is what sets the twenty-five-year window the admissions-rate projection above is fitted on.

Does it survive a year it has never seen? The model is fitted only on years UGA has filed, through Fall 2025. UGA has since announced its Class of 2030 result, which makes an honest out-of-sample test. PI is the prediction interval — the range the model said the answer should fall in:

Fall 2026Predicted90% PIActualResult
Applicants45,20538,068 – 53,681~51,600Inside
Admit rate36.8%29.2% – 44.5%~29.8%Inside

Both landed inside the band, which is the most this kind of test can say for a model. The rate still came in 7 points below the central estimate, near the bottom of the interval, so the bias has a direction worth stating: this model under-predicts how quickly UGA is tightening. Treat the projected rates as the optimistic case.

Every year, as UGA published it — 66 filed, plus the announced class of 2030
FallAppliedAdmittedRateTurned awayMean SATvs nationalSource
19602,5932,19284.5%401FB 1994
19612,8352,22078.3%615890−79FB 1989
19623,0942,49580.6%599924−47FB 1994
19633,5672,83079.3%737FB 1994
19643,9813,24681.5%735FB 1999
19654,6853,87382.7%812FB 2000
19665,1504,01377.9%1,1371020+53FB 2001
19675,2073,69170.9%1,5161012+54FB 2002
19685,6724,41677.9%1,2561032+74FB 2002
19695,4844,15975.8%1,3251046+90FB 2002
19705,6004,29176.6%1,3091038+90FB 2002
19715,1763,92375.8%1,2531039+96FB 2002
19725,4424,21177.4%1,2311019+82FB 2002
19735,1413,92576.3%1,2161012+86FB 2002
19745,3234,49384.4%830986+62FB 2002
19755,1914,18380.6%1,008994+88FB 2002
19765,5904,29276.8%1,298999+96FB 2002
19775,5304,29077.6%1,2401013+114FB 2002
19786,1774,56473.9%1,6131022+125FB 2002
19797,0724,62365.4%2,4491028+134FB 2002
19806,7164,59768.4%2,1191026+136FB 2002
19817,4915,90678.8%1,5851000+110FB 2002
19827,3025,46074.8%1,8421026+133FB 2002
19837,5045,43072.4%2,0741017+124FB 2002
19847,8045,85675.0%1,9481001+104FB 2002
19858,7196,74377.3%1,9761012+106FB 2002
19869,7787,46076.3%2,3181016+110FB 2002
198710,0137,58175.7%2,4321028+122FB 2002
198810,8788,01573.7%2,8631033+129FB 2002
198910,6837,42269.5%3,2611048+145FB 2002
19909,5617,51378.6%2,0481045+145FB 2002
199110,5097,08867.4%3,4211060+164FB 2002
199211,3637,30464.3%4,0591078+179FB 2002
199311,2207,87170.2%3,3491086+184FB 2002
199412,3827,88563.7%4,4971084+182FB 2003
199513,4017,93559.2%5,4661109+199FB 2004
199612,9307,25956.1%5,671FB 2004
199711,5138,41173.1%3,102FB 2006
199812,7288,45966.5%4,269FB 2007
199913,4028,46663.2%4,936FB 2007
200012,8697,93261.6%4,9371202+183FB 2009
200110,9847,97672.6%3,0081207+187FB 2009
200212,7868,34065.2%4,4461215+195FB 2002
200311,8138,88575.2%2,9281212+186FB 2011
200413,2678,19761.8%5,0701233+207FB 2011
200512,3267,98264.8%4,3441241+213CDS 2005-06
200615,9249,21457.9%6,7101228+207FB 2014
200716,9999,27954.6%7,7201233+216FB 2014
200817,3269,62555.6%7,7011237+220FB 2014
200917,8869,58453.6%8,3021242+226FB 2014
201017,73010,33158.3%7,3991239+222FB 2019
201117,82311,17062.7%6,6531226+215FB 2019
201218,73510,44955.8%8,2861244+234FB 2019
201320,34011,05454.3%9,2861249+239FB 2024
201420,91811,65555.7%9,2631252+242FB 2024
201521,94511,60452.9%10,341FB 2024
201622,69412,23253.9%10,462FB 2024
201724,16513,05254.0%11,113FB 2024
201826,02712,65948.6%13,368FB 2024
201929,06513,26145.6%15,8041334+275FB 2024
202028,02413,75049.1%14,2741318+267FB 2024
202139,09015,68940.1%23,4011330+280FB 2024
202239,35416,73042.5%22,6241305+255FB 2024
202343,41516,14837.2%27,2671307+279FB 2024
202442,43616,09237.9%26,3441304+280FB 2024
202547,46416,25734.3%31,207CDS 2025-26
2026~51,600~15,40029.8%~36,200Announced

If your student is aiming at Georgia, I’ll tell you where they actually stand.

Send an inquiry
Call Text