Can we predict how NDA will perform?
Let's try a few approaches and see what works—and what doesn't.
Can we use past Assembly elections?
Let's start simple—look at how NDA performed in past Assembly elections.
| Year | Election | NDA Vote Share | Turnout |
|---|---|---|---|
| 2011 | Assembly | 6.0% | 75.3% |
| 2016 | Assembly | 15.0% | 77.5% |
| 2021 | Assembly | 12.4% | 76.0% |
Seems straightforward. But wait...
- Only 3 data points—that's a tiny sample
- Ignores what happened in LS and Local elections
What about Local Body elections?
They're more recent. Maybe they're a better signal?
| Year | Election | NDA Vote Share |
|---|---|---|
| 2015 | Local Body | 13.3% |
| 2020 | Local Body | 15.0% |
| 2025 | Local Body | 16.0% |
Closer to 2026, but...
- Local elections measure ground-level organization, not state voting behavior
- Turnout patterns are different
- Local issues dominate—state and national factors are muted
What if we look at all elections?
Each election type tells a different part of the story.
Three different numbers. LS says 19.4%, Local says 16.0%, Assembly says 12.4%.
We can't just average them—they need to be translated to Assembly terms.
How do we combine these signals?
This is where assumptions come in. Adjust them to see how sensitive the forecast is.
But how confident should we be?
16.3% looks precise—but elections aren't.
The real question isn't "what's the number?"—it's "how likely are we to cross meaningful thresholds?"
The map of possibilities
Instead of one number, let's think in chances.
So what did we learn?
"This isn't a prediction. It's a map of plausible futures."
- 0% chance of ≥16% — less than even odds, but definitely possible
- 100% chance of falling short — that's real too
- The forecast is only as good as the assumptions — scroll back up and try different ones
The goal isn't to find "the right answer"—it's to understand what drives the range of possibilities.