"You Used to Call Me on My Cell Phone"
In Drake's *Hotline Bling* (2015), the question is whether to reach out. The answer, per the song, is complicated. Per this calculator, it's a probabilistic decision based on three inputs and one absolute rule (your sober friends are right).
How to Use This Calculator
Hours since last contact. Default 18.
Drinks consumed in the last 3 hours. Default 0. Each drink penalizes the call-back signal because alcohol inflates confidence.
Friend group verdict. Yes / no / mixed. Default mixed.
The Formula
A simple weighted score. If your friends say no AND you've had >2 drinks, the calculator hard-rules NO. Otherwise:
signal = (24 โ hours) ร 2 โ (drinks ร 15) + friend_score
decision = signal > 0 ? "Call" : "Don't call"
Frequently Asked Questions
Should I really not call them back?
Listen to your friends.
Did Drake actually want them to call?
Read his lyrics carefully. He misses them. He also implies they've changed. Both can be true.
Is double-texting OK?
Different calculator.
What's the realistic "leave it" threshold from the dating-cadence data?
Pew and Match.com's 2024 dating surveys show that the median time-to-reply for a low-stakes text in an existing dating relationship is about 2โ6 hours during waking hours, and under 24 hours overall. Once you cross 48 hours of silence with no plausible logistical explanation (travel, work crisis, illness), the conversation has effectively ended for most people. The calculator's decay weighting reflects that: it doesn't punish a 12-hour gap, but it heavily penalizes a 60-hour gap.
Why does alcohol get such a steep penalty?
Because the regret rate on drunk texts is empirically high. Survey work on intoxicated communication consistently finds that people overestimate the warmth of their message and underestimate the recipient's read of it. The โ15 per drink is a deliberately harsh weight; the calculator is structured to bias you toward not sending the text under the influence, because the wins from sending are small and the losses are large.
Should friend consensus really be the dominant input?
Yes. The friend-group verdict is the only input here that synthesizes context the calculator can't see: prior breakup pattern, the other person's current dating status, whether your last conversation ended badly. The formula gives it weight because in real-world studies, the single best predictor of whether reaching out to an ex goes well is whether mutual friends are encouraging.
A Worked Example
It's been 36 hours since your last text. You've had three drinks. Your friend group is split โ one says go for it, two say absolutely not. Default mixed โ friend score 0.
The combination of late-night timing, alcohol, and non-supportive friends is exactly the configuration that produces tomorrow's regret. The calculator doesn't moralize โ it just runs the weights and tells you the answer your sober self would have given.
Behind the Numbers
"Hotline Bling" debuted in 2015 and became one of the most-meme'd music videos of the decade โ the choreography by Director X is its own cultural artifact. The song's actual content is a meditation on whether a former partner has changed in ways that exclude the speaker; the calculator is a deliberately reductive take on a song that is, lyrically, more ambivalent than its "you used to call me" hook suggests.
The numeric weights here are not a scientific model. They're calibrated to give an honest answer in the most common failure mode of the question (late-night, alcohol-driven impulse) while not overcorrecting in the common success mode (sober, mid-day, supportive friends). The hard-rule cutoff for friends-say-no + 2+ drinks is the only branch that overrides the linear math.