Spot the Lie

Four misleading charts — can you see what’s wrong?

Each chart below uses real data. Each chart is misleading. For each one, ask yourself: what impression does this give? Then: what’s wrong with it?

Chart 1: “Bioenergy is surging”

Show code

ggplot(elec, aes(x = year, y = bioenergy_twh)) +
  geom_line(linewidth = 1.2) +
  geom_point(size = 2) +
  coord_cartesian(ylim = c(33, 42)) +
  labs(x = "Year", y = "Generation (TWh)",
       title = "UK bioenergy generation is surging!",
       caption = "Source: DUKES 2025") +
  theme_minimal(base_size = 14)

Line chart of UK bioenergy generation with a y-axis starting at 33 TWh instead of zero. The trend appears to show dramatic growth, but the actual range is only about 35 to 40 TWh. — Figure 1: UK bioenergy electricity generation (TWh)

What’s wrong?

Truncated y-axis. The y-axis starts at 33, not 0. This makes the variation from ~35 to ~40 TWh fill the entire plot, giving the impression of dramatic growth. In reality, bioenergy has been roughly flat since 2020, fluctuating within a narrow band.

The honest version would start the y-axis at 0, showing that the recent changes are small relative to the total.

Show code

ggplot(elec, aes(x = year, y = bioenergy_twh)) +
  geom_line(linewidth = 1.2) +
  geom_point(size = 2) +
  coord_cartesian(ylim = c(0, 45)) +
  labs(x = "Year", y = "Generation (TWh)",
       title = "UK bioenergy generation",
       caption = "Source: DUKES 2025") +
  theme_minimal(base_size = 14)

Corrected line chart with y-axis starting at zero. The bioenergy trend is roughly flat since 2020, within a narrow band around 35 to 40 TWh.

Same data. Very different impression.

Chart 2: “Biomass has replaced coal”

Show code

elec_subset <- elec[elec$year >= 2012 & elec$year <= 2020, ]

ggplot(elec_subset, aes(x = year)) +
  geom_line(aes(y = bioenergy_twh, colour = "Bioenergy",
                linetype = "Bioenergy"), linewidth = 1.2) +
  geom_line(aes(y = coal_twh, colour = "Coal",
                linetype = "Coal"), linewidth = 1.2) +
  scale_colour_manual(values = c("Bioenergy" = "forestgreen",
                                  "Coal" = "grey40")) +
  labs(x = "Year", y = "Generation (TWh)",
       colour = NULL, linetype = NULL,
       title = "Clean bioenergy is replacing dirty coal",
       caption = "Source: DUKES 2025") +
  theme_minimal(base_size = 14) +
  theme(legend.position = "top")

Line chart showing only bioenergy and coal from 2012 to 2020. Bioenergy rises while coal falls, creating the impression that biomass replaced coal. Other fuels are excluded. — Figure 2: UK electricity generation: selected fuels

What’s wrong?

Two problems:

1. Cherry-picked date range. Only 2012–2020 is shown — the period when biomass grew fastest and coal fell fastest. The full 2000–2024 series tells a more complete story: coal’s decline started before biomass grew, and most of the replacement came from gas and wind, not biomass.

2. Only two fuels shown. By excluding gas, wind, nuclear, and solar, the chart creates the false impression that biomass specifically replaced coal. In reality, wind grew from 8 TWh to 83 TWh over this period — dwarfing biomass’s contribution.

Show code

ggplot(elec, aes(x = year)) +
  geom_line(aes(y = coal_twh, colour = "Coal", linetype = "Coal"),
            linewidth = 1) +
  geom_line(aes(y = gas_twh, colour = "Gas", linetype = "Gas"),
            linewidth = 1) +
  geom_line(aes(y = wind_twh, colour = "Wind", linetype = "Wind"),
            linewidth = 1) +
  geom_line(aes(y = bioenergy_twh, colour = "Bioenergy",
                linetype = "Bioenergy"), linewidth = 1) +
  labs(x = "Year", y = "Generation (TWh)", colour = "Fuel",
       linetype = "Fuel",
       title = "UK electricity generation by fuel (2000–2024)",
       caption = "Source: DUKES 2025") +
  theme_minimal(base_size = 14)

Corrected line chart showing coal, gas, wind, and bioenergy from 2000 to 2024. Wind grew from near zero to about 83 TWh, dwarfing bioenergy's contribution. Coal declined and gas dominates.

Biomass is part of the picture — but a small part.

Chart 3: “Biomass is the cleanest fuel”

Show code

# Asymmetric comparison: official for biomass, supply-chain for coal
asym <- data.frame(
  fuel = c("Biomass", "Coal", "Gas"),
  co2 = c(0, 990, 410),
  label = c("Official\n(UNFCCC)", "With supply\nchain", "With supply\nchain")
)
asym$fuel <- factor(asym$fuel, levels = c("Biomass", "Gas", "Coal"))

ggplot(asym, aes(x = fuel, y = co2)) +
  geom_col(fill = c("forestgreen", "steelblue", "grey40"),
           width = 0.6) +
  geom_text(aes(label = paste0(co2, "\nkg/MWh")),
            vjust = -0.3, size = 4.5) +
  coord_cartesian(ylim = c(0, 1150)) +
  labs(x = NULL, y = "CO₂ (kg per MWh)",
       title = "Biomass produces zero emissions",
       caption = "Sources: UNFCCC, DESNZ") +
  theme_minimal(base_size = 14)

Bar chart showing CO2 per MWh. Biomass at 0 (using official UNFCCC accounting), gas at 410, and coal at 990. The biomass figure uses different accounting rules from the other fuels. — Figure 3: CO₂ emissions by fuel — official figures

What’s wrong?

Asymmetric accounting. Biomass uses the “official” emission factor (0 kg CO₂/MWh — the UNFCCC convention that ignores combustion emissions). But coal and gas use supply-chain figures (990 and 410), which include mining, processing, and transport. This is an apples-to-oranges comparison.

If you apply the same accounting rules to all fuels, the picture changes dramatically:

Fuel	Official	With supply chain
Biomass	0	390
Coal	910	990
Gas	360	410

Under consistent supply-chain accounting, biomass emits 390 kg/MWh — comparable to gas, not zero.

The general principle: When comparing numbers, check that the same rules were applied to all of them. Asymmetric standards are one of the hardest tricks to spot because each individual number is correct.

Chart 4: “The UK imports a trivial amount of wood”

Show code

# Misleading scale comparison
comp <- data.frame(
  category = c("Wood pellets", "All goods"),
  value = c(9.3, 550000),
  unit = c("kt", "kt")
)
comp$category <- factor(comp$category,
                         levels = c("Wood pellets", "All goods"))

ggplot(comp, aes(x = category, y = value)) +
  geom_col(width = 0.5) +
  labs(x = NULL, y = "Import volume (thousand tonnes)",
       title = "Wood pellet imports in context",
       caption = "Sources: Forest Research, ONS") +
  theme_minimal(base_size = 14)

Bar chart comparing wood pellet imports (9.3 kt) to total UK goods imports (550,000 kt). The pellet bar is invisible next to the total, making imports seem trivial — but the comparator is meaningless. — Figure 4: UK wood pellet imports vs total goods imports

What’s wrong?

Meaningless comparator. Comparing 9.3 million tonnes of wood pellets to the UK’s total goods imports (~550 Mt) makes pellets look like a rounding error. But “all goods” includes crude oil, iron ore, vehicles, food — everything. It’s not a useful denominator.

The pellet bar is effectively invisible. The visual impression is “nothing to see here” — which might be the point.

A more useful comparison would be:

9.3 Mt of pellets is roughly 2% of all UK biomass consumption
It requires approximately 23 million trees’ worth of wood per year (depending on species and age)
It is shipped an average of ~5,500 km, mostly from the southeastern United States
It arrives at a small number of UK ports, primarily for one power station (Drax)

Context matters. The question isn’t whether pellets are a large fraction of all imports — it’s whether the supply chain is compatible with the “carbon-neutral” claim.

Discussion

For each chart, ask:

What is the chart trying to make me feel? (Before analysing the data, notice your first impression.)
What question would reveal the deception? (“Compared to what?” / “Why this date range?” / “Are the same rules applied to all bars?”)
What would the honest version look like? (What would you change to make the chart fair?)

Key point: None of these charts contain false data. Every number is real and sourced. The deception is in the choices: which data to show, which to omit, where to start the axis, and what to compare. These are the same choices you’ll make in your own briefing — so make them deliberately and honestly.